An environment self-adaption based temperature and humidity anomaly prediction and prevention method and system

Abstract

This invention discloses a method and system for predicting and preventing abnormal temperature and humidity based on environmental adaptation, belonging to the field of abnormal monitoring technology. It collects historical data through a temperature and humidity detection module to construct a dataset containing temperature and humidity data. Multidimensional clustering analysis is performed based on the equipment's operating cycle and the weight of the items to be dried. Benchmark temperature and humidity change curves are established for different weight categories, forming a standard reference model for normal equipment operation. Real-time temperature and humidity data are collected to construct dynamic change curves, which are compared with the benchmark curves for the corresponding weight scenarios to achieve accurate identification of abnormal states. The system automatically determines the type of sensor fault and triggers an alarm mechanism, while simultaneously activating backup sensors to ensure continuous monitoring. If mixed drift characteristics are observed, a regular alarm is output to prompt maintenance. The system employs a dual-sensor redundancy design, combined with a weight-adaptive benchmark curve matching strategy, achieving an intelligent transformation from passive alarm to proactive prevention.

Description

Technical Field

[0001] This invention belongs to the field of anomaly monitoring technology, specifically, it relates to a method and system for predicting and preventing temperature and humidity anomalies based on environmental adaptation. Background Technology

[0002] In temperature and humidity control devices such as washing machines and dryers that rely on temperature and humidity detection data for operational logic judgment, the temperature and humidity sensor module is the core sensing component, and the accuracy of its data directly determines the operating effect of the device.

[0003] Existing technologies generally lack detailed consideration of the operating cycle and the weight of the items to be dried. Items of different weights have significantly different heat and moisture load characteristics. Lightweight items heat up quickly and release moisture rapidly, while heavyweight items require longer preheating times and a more stable humidity decrease curve. Existing technologies often use uniform and fixed upper and lower limits of temperature and humidity as the criteria for judging anomalies, without considering the normal changes that should occur at different stages of the operating cycle, and without distinguishing the weight class of the items being dried. For example, in the early stages of drying, when the environment is not yet stable and the temperature is rising from room temperature to the operating temperature, existing technologies are very prone to frequently triggering false alarms due to the instantaneous temperature exceeding the fixed threshold. In the middle and later stages of drying, when the temperature and humidity should decrease slowly, existing technologies may miss the real anomalies because the actual values ​​deviate from the static threshold. This gradual failure is often ignored under fixed thresholds and can only be detected when the device completely fails. The existing technology has a relatively simple anomaly judgment logic, which mostly uses static thresholds or sudden change detection of time difference values. It cannot capture the complex coupling and drift phenomena between temperature and humidity change curves, and it is difficult to effectively identify compound faults. Based on the parameters set by engineers based on experience, the system has extremely poor versatility and robustness. Secondly, the existing technology lacks an active prevention mechanism. The accumulation of latent anomalies can easily lead to explicit faults, which in turn cause abnormal equipment operation.

[0004] To address the aforementioned problems, this invention proposes a method and system for predicting and preventing abnormal temperature and humidity based on environmental adaptation. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for predicting and preventing temperature and humidity anomalies based on environmental adaptation, which solves the problems of existing technologies being unable to adaptively distinguish temperature and humidity anomalies of different weights and cycles, as well as the lack of sensor redundancy and fault tolerance.

[0006] The objective of this invention can be achieved through the following technical solutions: A method for predicting and preventing abnormal temperature and humidity based on environmental adaptation, the method comprising: Step 1: Monitor the target temperature and humidity control device based on the temperature and humidity detection module, collect the temperature and humidity data associated with the target temperature and humidity control device in the historical period, align them by time, and form a historical temperature and humidity data set. Step 2: Using the operating cycle as the interval, perform a partitioning operation on the historical temperature and humidity data set to determine the interval historical temperature data sequence and the interval historical humidity data sequence; Clustering is performed based on the weight of the items to be dried in each operating cycle to lock the secondary interval historical temperature data sequence and the secondary interval historical humidity data sequence of the items to be dried under different weights. Step 3: Construct the temperature change curve and humidity change curve associated with the secondary interval historical temperature data sequence and the secondary interval historical humidity data sequence of all items to be dried of the same weight. Perform a fitting operation to lock the reference temperature change curve and reference humidity change curve associated with the target temperature and humidity control equipment under the corresponding weight during the operating cycle. Step 4: Determine the weight of the items to be dried in the current target temperature and humidity control equipment, construct real-time temperature change curves and real-time humidity change curves, combine them with the corresponding weight's reference temperature change curves and reference humidity change curves to perform a verification operation, identify abnormal changes, and output an abnormal alarm.

[0007] As a further aspect of the present invention, the specific method for constructing the historical temperature and humidity data set in step one is as follows: Identify the target temperature and humidity control equipment, denoted as Q; Based on the temperature and humidity detection module, the temperature and humidity data of the target temperature and humidity control device Q at any time during any operating state within the historical period are obtained and constructed as historical temperature data sequences H1, H2, ..., Hj and historical humidity data sequences W1, W2, ..., Wj, respectively. Here, j is the total number of durations of the operating state, which is determined by the temperature and humidity scheduling program built into the temperature and humidity control device of the target temperature and humidity control device Q specification. Hi corresponds to Wi, i is the counting index, and 1≤i≤j. The historical period refers to a past time period, the specific duration of which is determined by the operator. The historical temperature data sequence H1, H2, ..., Hj and the historical humidity data sequence W1, W2, ..., Wj are combined and denoted as the associated temperature and humidity data combination D; Similarly, acquire temperature and humidity data of several temperature and humidity control devices of the same specifications as the target temperature and humidity control device Q when they are in operation during the historical period, and form a historical temperature data sequence, a historical humidity data sequence, and a combination of related temperature and humidity data. All the associated temperature and humidity data combinations constructed are summarized and denoted as the historical temperature and humidity data set {D1,D2,...,Dm}, where m is the total number of associated temperature and humidity data combinations.

[0008] As a further aspect of the present invention, in step one, the temperature and humidity detection module includes a first temperature sensor and a second temperature sensor, as well as a first humidity sensor and a second humidity sensor. The first temperature sensor and the first humidity sensor are the main sensors. When the first temperature sensor and the first humidity sensor are abnormal, the system switches to the second temperature sensor and the second humidity sensor to perform data acquisition.

[0009] As a further aspect of the present invention, the specific method for determining the interval historical temperature data sequence and the interval historical humidity data sequence in step two is as follows: Obtain the historical temperature and humidity data set {D1,D2,...,Dm}; The statistical analysis includes all temperature and humidity scheduling programs for the target temperature and humidity control equipment of specification Q, denoted as CX1,CX2,...,CXu, where u is the total number of temperature and humidity scheduling programs. Get all associated temperature and humidity data combinations of all temperature and humidity control devices under any temperature and humidity scheduling program CXo during the operating cycle, and summarize them as the historical temperature and humidity data subset {Do1,Do2,...,Dor}, where r is the total number of associated temperature and humidity data combinations in the historical temperature and humidity data subset, 0≤r≤m, and o is the counting index, with a value range of 1≤o≤u; Separate the historical temperature data sequence and historical humidity data sequence corresponding to all associated temperature and humidity data combinations in the historical temperature and humidity data subset {Do1,Do2,...,Dor} and label them as interval historical temperature data sequence and interval historical humidity data sequence. Similarly, determine the subset of historical temperature and humidity data associated with the operating cycle corresponding to all temperature and humidity scheduling programs, as well as the interval historical temperature data sequence and interval historical humidity data sequence.

[0010] As a further aspect of the present invention, the specific method for performing clustering operations in step two, based on the weight of the items to be dried in each operating cycle, is as follows: Get the historical temperature and humidity data subset {Do1,Do2,...,Dor} associated with any temperature and humidity scheduler CXo; Simultaneously extract the weights of r associated temperature and humidity data combinations within the corresponding operating cycle, denoted as Go1, Go2, ..., Gor. Arrange Go1, Go2, ..., Gor in ascending order, and denote the weight sequence Go1', Go2', ..., Gor'. Using Go1' as the baseline, calculate the difference rate CY1 between Go1' and Go2', and compare it with the preset difference rate threshold CY. yu Comparison, if CY1≤CY yu Extract Go2', cluster it with Go1' into category F1, and continue to perform calibration operations with Go1' on subsequent weights; If there exists any weight, the difference rate between Goe' and Go1' is greater than CY yu Then, Goe' is used as the new baseline, and the unextracted weights are traversed. Clustering operations are performed to cluster into category F2, where e is the counting index and 1≤e≤r. This process continues until all weights in the weight sequence Go1',Go2',...,Gor' are clustered. The total number of clusters is counted and denoted as v. The historical temperature and humidity data subset {Do1,Do2,...,Dor} is clustered based on the weight of the items to be dried and their corresponding categories F1,F2,...,Fv. All interval historical temperature data sequences and interval historical humidity data sequences corresponding to any associated temperature and humidity data combinations in any category Fb are labeled as secondary interval historical temperature data sequences and secondary interval historical humidity data sequences, where 1≤b≤v.

[0011] As a further aspect of the present invention, the specific method for locking the reference temperature change curve and reference humidity change curve associated with the target temperature and humidity control device under the corresponding weight during the operating cycle in step three is as follows: Determine any associated temperature and humidity data combination Doe, which is classified into any category Fb, and its corresponding secondary interval historical temperature data sequence He1',He2',...,Hej' and secondary interval historical humidity data sequence We1',We2',...,Wej'; Construct a two-dimensional coordinate system XY1 with time as the horizontal axis, humidity as the positive half-axis, and temperature as the negative half-axis, and set the scale value of the negative half-axis to positive. The horizontal axis spans j time points. Align the historical temperature data sequence He1',He2',...,Hej' of the secondary interval with the historical humidity data sequence We1',We2',...,Wej' of the secondary interval with the horizontal axis scale, and plot them in the two-dimensional coordinate system XY1 in the form of data points to obtain two sets of data points. The data points are then fitted with curves and recorded as the temperature change curve and humidity change curve corresponding to the associated temperature and humidity data combination Doe, respectively. Similarly, determine the temperature change curves and humidity change curves corresponding to all associated temperature and humidity data combinations in category Fb; Fit all temperature change curves to obtain the reference temperature change curve S1. Similarly, by fitting all humidity change curves, the baseline humidity change curve S2 is obtained; As described above, determine the reference temperature change curve and reference humidity change curve for any weight scenario of any item to be dried under any temperature and humidity scheduling program.

[0012] As a further aspect of the present invention, in step four, the specific method for determining the weight of the items to be dried in the current target temperature and humidity control device and constructing the real-time temperature change curve and the real-time humidity change curve is as follows: When the target temperature and humidity control device Q is in the ready-to-operate state, obtain the temperature and humidity scheduling program of the target temperature and humidity control device Q and the weight of the items to be dried. Based on the current target temperature and humidity control device Q's temperature and humidity scheduling program and the weight lock of the items to be dried, the corresponding reference temperature change curve S1 and reference humidity change curve S2 are used. When the target temperature and humidity control device Q is in operation, it monitors and collects temperature and humidity data in real time, and constructs real-time temperature change curve S3 and real-time humidity change curve S4 according to the time sequence.

[0013] As a further aspect of the present invention, the specific method for determining abnormal changes and outputting abnormal alarms in step four is as follows: The real-time temperature change curve S3 and the real-time humidity change curve S4 are plotted in the two-dimensional coordinate system XY1 where the reference temperature change curve S1 and the reference humidity change curve S2 are located. Using the preset total number of time points k as the dividing interval, starting from the 0 mark of the horizontal axis of the two-dimensional coordinate system XY1, the horizontal axis is divided to determine the first dividing interval segment. Then, the straight lines perpendicular to the horizontal axis and parallel to the vertical axis are drawn through the 0 mark and the k mark on the dividing interval segment, respectively, and are recorded as the first dividing line L1 and the second dividing line L2. Calculate the area of ​​the closed interval formed by S1, S3, L1, and L2, and denote it as the temperature difference characteristic TZH; Calculate the area of ​​the closed interval formed by S2, S4, L1, and L2, and denote it as the humidity difference characteristic TZW; Extract the preset temperature difference threshold H yu Humidity difference threshold W yu ; The temperature difference feature TZH and humidity difference feature TZW are respectively compared with the temperature difference threshold H. yu Humidity difference threshold W yu Perform the comparison; If TZH≥H yu If the first temperature sensor is found to be malfunctioning, an alarm will be output and the system will switch to the second temperature sensor. If TZW≥W yuIf the first humidity sensor is found to be malfunctioning, an alarm will be output and the system will switch to the second humidity sensor. If TZH < H yu And TZW < W yu The mixed drift characteristic SUM is calculated using SUM=TZH+TZW; The correlation anomaly feature SUM is compared with a preset mixed drift threshold SUM. yu Perform the comparison; If SUM≥SUM yu If the first humidity sensor and the first temperature sensor are found to be experiencing mixed drift, a standard alarm will be output, and the operator will be notified to perform maintenance. Conversely, continuous monitoring is required. Similarly, monitor all segmented intervals, identify abnormal changes, and output anomaly alerts.

[0014] An environmentally adaptive temperature and humidity anomaly prediction and prevention system, the system comprising: The all-domain sensing and acquisition terminal monitors the target temperature and humidity control equipment based on the temperature and humidity detection module, and collects the temperature and humidity data associated with the target temperature and humidity control equipment in the historical period, aligns them by time, and forms a historical temperature and humidity data set. The spatiotemporal clustering decomposition end performs a partitioning operation on the historical temperature and humidity data set with the running cycle as the interval, and determines the interval historical temperature data sequence and interval historical humidity data sequence. Clustering is performed based on the weight of the items to be dried in each operating cycle to lock the secondary interval historical temperature data sequence and the secondary interval historical humidity data sequence of the items to be dried under different weights. In the baseline fitting modeling end, the temperature change curve and humidity change curve associated with the secondary interval historical temperature data sequence and the secondary interval historical humidity data sequence of all items to be dried of the same weight are constructed. The fitting operation is performed to lock the reference temperature change curve and reference humidity change curve associated with the target temperature and humidity control equipment under the corresponding weight during the operating cycle. The dynamic deviation verification terminal determines the weight of the items to be dried in the current target temperature and humidity control equipment, constructs real-time temperature change curves and real-time humidity change curves, and performs verification operations by combining the corresponding weight's reference temperature change curves and reference humidity change curves to identify abnormal changes and output abnormal alarms.

[0015] The beneficial effects of this invention are: This invention collects historical temperature and humidity data and performs cluster analysis by associating it with the weight of the items to be dried. It constructs benchmark temperature and humidity change curves for different weight loads, enabling precise modeling of the operating status of the target temperature and humidity control equipment. It has environmental adaptability and can dynamically adjust the prediction benchmark according to the actual weight of the items to be dried, avoiding false alarms or missed alarms caused by load differences in traditional fixed threshold alarm methods. Secondly, by verifying with real-time change curves, it can identify abnormal trends in advance and issue alarms, thereby intervening before temperature and humidity runaway, effectively preventing problems such as uneven drying, equipment overload or energy waste, improving the stability of the drying process and product quality, and reducing equipment failure risk and maintenance costs. This invention ensures the accurate preservation of temperature and humidity correlation by combining the historical temperature and humidity sequences of the target device with those of devices of the same specifications according to time intervals, thus avoiding data fragmentation. At the same time, it introduces a redundant design of primary and backup temperature and humidity sensors, which automatically switches to the backup sensor when the primary sensor fails, effectively ensuring the continuity of data acquisition and system robustness, and reducing the risk of data loss due to single point of failure. This enables high-quality accumulation of historical data and enhances the fault tolerance of the sensor system, thereby improving the stability and accuracy of temperature and humidity control. This invention innovatively introduces the weight of the items to be dried as a clustering basis, improving the precision and intelligence of temperature and humidity control. Its beneficial effects include dynamic clustering based on weight difference rate, which adaptively separates drying batches of different weight levels, avoiding the bias caused by fixed grouping, and thus more accurately identifying typical temperature and humidity variation patterns within each weight range; it also uncovers potential correlations in historical data, making the extracted secondary interval historical temperature and humidity sequences highly targeted and representative, reducing the arbitrariness and workload of manual grouping, and improving data processing efficiency; and through deep binding of weight and temperature and humidity data, it provides a scientific basis for subsequent scheduling procedures. This invention constructs a two-dimensional coordinate system that simultaneously characterizes temperature and humidity, and performs quantitative analysis of the area difference between historical baseline curves and real-time monitoring curves. This allows for an intuitive and quantitative reflection of the deviation between the actual curve and the baseline curve, avoiding subjective judgment errors. Secondly, by setting thresholds for comparison, it can accurately distinguish between single sensor failures and mixed drift, and automatically trigger backup sensor switching or output regular alarms, ensuring the continuous operation capability of the equipment under critical anomalies and reducing the risk of false alarms and missed alarms caused by complex coupling anomalies. Attached Figure Description

[0016] The invention will now be further described with reference to the accompanying drawings.

[0017] Figure 1 This is a flowchart illustrating the method described in this invention; Figure 2 This is a schematic diagram of the system described in this invention. Detailed Implementation

[0018] 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. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] like Figure 1 as well as Figure 2 As shown, this application provides a method and system for predicting and preventing abnormal temperature and humidity based on environmental adaptation; As an embodiment 1 of this application, it specifically includes: The all-domain sensing and acquisition terminal monitors the target temperature and humidity control equipment based on the temperature and humidity detection module, and collects the temperature and humidity data associated with the target temperature and humidity control equipment in the historical period, aligns them by time, and forms a historical temperature and humidity data set. The spatiotemporal clustering decomposition end performs a partitioning operation on the historical temperature and humidity data set with the running cycle as the interval, and determines the interval historical temperature data sequence and interval historical humidity data sequence. Clustering is performed based on the weight of the items to be dried in each operating cycle to lock the secondary interval historical temperature data sequence and the secondary interval historical humidity data sequence of the items to be dried under different weights. In the baseline fitting modeling end, the temperature change curve and humidity change curve associated with the secondary interval historical temperature data sequence and the secondary interval historical humidity data sequence of all items to be dried of the same weight are constructed. The fitting operation is performed to lock the reference temperature change curve and reference humidity change curve associated with the target temperature and humidity control equipment under the corresponding weight during the operating cycle. The dynamic deviation verification terminal determines the weight of the items to be dried in the current target temperature and humidity control equipment, constructs real-time temperature change curves and real-time humidity change curves, and performs verification operations by combining the corresponding weight's reference temperature change curves and reference humidity change curves to identify abnormal changes and output abnormal alarms. Example 2

[0020] This embodiment, based on Embodiment 1, further discloses a method for constructing a combination of correlated temperature and humidity data and a historical temperature and humidity data set, specifically including the following: This invention provides a method and system for predicting and preventing abnormal temperature and humidity based on environmental adaptation. It is mainly applicable to temperature and humidity control equipment that relies on temperature and humidity detection data for operational logic judgment. For example, it may be unable to adaptively adjust the reference curve according to the weight of the items to be dried, or misjudgment or missed judgment may occur due to a single sensor failure, or there may be a lack of fitting and prediction mechanism based on historical data. The main purpose is to monitor sensor abnormalities in reverse through a data-driven approach. Therefore, in actual implementation, the temperature and humidity control equipment needs to be checked by the operator in advance and calibrated as normal temperature and humidity control equipment. It should also be noted that the temperature and humidity detection module described in this solution includes a first temperature sensor and a second temperature sensor, as well as a first humidity sensor and a second humidity sensor. The first temperature sensor and the first humidity sensor are the main sensors. When the first temperature sensor and the first humidity sensor are abnormal, the system switches to the second temperature sensor and the second humidity sensor to perform data acquisition.

[0021] When operators need to perform more redundant configurations, multiple backup sensors can be used, such as a third temperature sensor or a fourth temperature sensor.

[0022] First, the operator needs to identify an arbitrary target temperature and humidity control device to be monitored. For ease of distinction from the following text, the target temperature and humidity control device is marked as Q, and will be used as an example in the following text. First, data is collected from the target temperature and humidity control device Q using the temperature and humidity detection module, including temperature data and humidity data. All past temperature and humidity data will be stored in a pre-built database for backup. Acquire the temperature and humidity data of the target temperature and humidity control device Q at any operating state within the historical period and corresponding to the time (i.e., include the temperature and humidity data of the target temperature and humidity control device Q at all times during one operating state, and match the temperature data and humidity data at the same time). The temperature and humidity data are arranged in chronological order to obtain historical temperature data sequences H1, H2, ..., Hj and historical humidity data sequences W1, W2, ..., Wj. The historical temperature data sequences H1, H2, ..., Hj and historical humidity data sequences W1, W2, ..., Wj are aligned with time, where j is the total number of durations of the target temperature and humidity control device Q in the current operating state, Hi corresponds to Wi, i is the counting index, and 1≤i≤j; The total number of durations is determined by the built-in temperature and humidity scheduling program of the target temperature and humidity control device Q specification. For example, if a certain temperature and humidity scheduling program is defined as a running state process with a duration of 3000 durations, then j=3000. The actual duration of the duration is determined by the operator, for example, a duration equal to 1 second or a duration equal to 5 seconds.

[0023] Based on the determined historical temperature data sequence H1, H2, ..., Hj and historical humidity data sequence W1, W2, ..., Wj, they are aligned according to time and combined to form a set of associated temperature and humidity data combination D. The associated temperature and humidity data combination D can be reverse-split into historical temperature data sequence H1, H2, ..., Hj and historical humidity data sequence W1, W2, ..., Wj. The associated temperature and humidity data combination D represents the evolution trajectory of temperature and humidity of the target temperature and humidity control device Q over time in this operating state.

[0024] Following the above method, temperature and humidity data of several temperature and humidity control devices of the same specifications as the target temperature and humidity control device Q during the historical period are extracted. Based on the temperature and humidity data of several temperature and humidity control devices, historical temperature data sequences, historical humidity data sequences, and associated temperature and humidity data combinations are constructed. Then, all the associated temperature and humidity data are combined and summarized in a random arrangement. The resulting data set is denoted as the historical temperature and humidity data set {D1, D2, ..., Dm}, where m is the total number of associated temperature and humidity data combinations. Each element in the historical temperature and humidity data set is a set of associated temperature and humidity data combinations, and a set of associated temperature and humidity data combinations corresponds to a historical temperature data sequence and a historical humidity data sequence. Example 3

[0025] This embodiment further discloses a method for segmenting historical temperature and humidity data based on weight clustering, building upon Embodiment 2. Specifically, it includes the following: First, obtain the historical temperature and humidity data set {D1,D2,...,Dm} constructed in Example 2; Furthermore, all temperature and humidity scheduling programs of the target temperature and humidity control equipment Q specification are obtained and denoted as CX1, CX2, ..., CXu, where u is the total number of temperature and humidity scheduling programs. Based on the different temperature and humidity control logic corresponding to different temperature and humidity scheduling programs, they are first classified according to scheduling programs to facilitate subsequent targeted analysis. At the same time, data mixing of different control strategies is avoided, and the semantic rationality of grouping is improved. Obtain any one temperature and humidity scheduler CXo from CX1, CX2, ..., CXu, and extract all associated temperature and humidity data combinations that operate with the temperature and humidity scheduler CXo within the running cycle from the historical temperature and humidity data set {D1, D2, ..., Dm}. Then, perform a summary operation to obtain the historical temperature and humidity data subset {Do1, Do2, ..., Dor}, where r is the total number of associated temperature and humidity data combinations in the historical temperature and humidity data subset, 0 ≤ r ≤ m, o is the counting index, and the value range is 1 ≤ o ≤ u, representing the temperature and humidity scheduler CXo. Next, all associated temperature and humidity data combinations are obtained from the historical temperature and humidity data subset {Do1,Do2,...,Dor} and separated to obtain all historical temperature data sequences and historical humidity data sequences in the historical temperature and humidity data subset {Do1,Do2,...,Dor}. These historical temperature data sequences and historical humidity data sequences are then labeled as interval historical temperature data sequences and interval historical humidity data sequences, representing data sequences divided into intervals based on the operating cycle.

[0026] Following the above method, all temperature and humidity scheduling programs are processed in the same way to obtain the historical temperature and humidity data subsets associated with each temperature and humidity scheduling program within the corresponding operating cycle, as well as the interval historical temperature data sequence and interval historical humidity data sequence.

[0027] Then, clustering is performed based on the weight of the items to be dried in each operating cycle, as follows: Continuing from the above, obtain the historical temperature and humidity data subset {Do1,Do2,...,Dor} associated with any temperature and humidity scheduling program CXo; Extract the weights of the items to be dried in the corresponding operating cycle from Do1 to Dor (determined by the weighing device integrated inside the temperature and humidity control equipment), totaling r weights, and align them with the historical temperature and humidity data subset {Do1,Do2,...,Dor} as Go1,Go2,...,Gor; Then sort Go1, Go2, ..., Gor in ascending order of their values, and denote the weight sequence as Go1', Go2', ..., Gor'; Next, traverse in the order of Go1' to Gor', using the first weight Go1' as the baseline, and calculate the difference rate between the subsequent weights and this baseline. A specific example will illustrate this: The difference rate between Go1' and Go2' is calculated as follows, denoted as CY1, where CY1 = |Go2' - Go1'| / (Go1' + ε), and ε is a very small positive number to prevent division by zero from causing anomalies. Compare the calculated difference rate CY1 with the preset difference rate threshold CY yu Perform a comparison; if the comparison result is CY1≤CY yu If so, extract Go2' and cluster Go2' and Go1' into the same category, denoted as category F1, and continue to perform calibration operations with Go1' on subsequent weights; If, in the above process, there exists any weight Goe' and Go1' whose difference rate is greater than CY. yu Then, Goe' is used as the new baseline, and based on the above steps, the unextracted weights are traversed and clustering operations are performed to cluster into category F2, where e is the counting index, 1≤e≤r; Repeat the above steps until all weights in the weight sequence Go1',Go2',...,Gor' have been clustered and classified into their respective categories; Next, count the total number of categories after the clustering operation, and label it as v. Then, assign the historical temperature and humidity data subset {Do1,Do2,...,Dor} to v categories: F1,F2,...,Fv according to the weight of the items to be dried, thus completing the clustering division. Next, obtain the interval historical temperature data sequence and interval historical humidity data sequence corresponding to all associated temperature and humidity data combinations in any category Fb, and mark them as secondary interval historical temperature data sequence and secondary interval historical humidity data sequence, respectively. Here, b is the counting index, and the value range is 1≤b≤v. In this way, the mapping relationship between weight interval and temperature and humidity pattern is established. Example 4

[0028] This embodiment further discloses an anomaly detection method for temperature and humidity sensors based on curve area comparison, building upon Embodiment 3. Specifically, it includes the following: Based on the content described in Example 3, extract any associated temperature and humidity data combination Doe that is classified into any category Fb, as well as the corresponding secondary interval historical temperature data sequence He1', He2', ..., Hej' and secondary interval historical humidity data sequence We1', We2', ..., Wej'. Construct a two-dimensional coordinate system XY1 with temperature and humidity coaxial. Specifically, the time line is used as the horizontal axis, the positive half of the vertical axis is humidity, and the negative half of the vertical axis is temperature. Simultaneously, the scale value of the negative half of the vertical axis is set to positive. It should be noted that the horizontal axis of the two-dimensional coordinate system XY1 spans j time points. Align the historical temperature data sequence He1',He2',...,Hej' of the secondary interval with the horizontal axis scale, and plot the j temperature data points in the two-dimensional coordinate system XY1 as data points to obtain a set of data points. Then, fit the data to a curve and denote it as the temperature change curve corresponding to the associated temperature and humidity data combination Doe. Similarly, based on the historical humidity data sequence of the quadratic interval We1',We2',...,Wej', construct the humidity change curve corresponding to the associated temperature and humidity data combination Doe; Then, following the steps above, determine the temperature change curves and humidity change curves corresponding to all associated temperature and humidity data combinations in category Fb. Fitting is performed on all temperature change curves and all humidity change curves respectively. The fitting operation can be based on a machine learning model, and finally the baseline temperature change curve S1 and the baseline humidity change curve S2 representing the temperature change of category Fb are obtained. By analogy, the baseline temperature change curve and baseline humidity change curve can be determined for any weight scenario (category) of the item to be dried under all temperature and humidity scheduling program scenarios. It should be noted that all the reference temperature and humidity change curves obtained above will be persistently stored. At the same time, the reference temperature and humidity change curves will be updated periodically based on newly added historical temperature and humidity data.

[0029] When the target temperature and humidity control device Q is in the ready-to-run state (the state about to run), first obtain the temperature and humidity scheduling program of the target temperature and humidity control device Q and the weight of the items to be dried inside the target temperature and humidity control device Q. Based on the current target temperature and humidity control device Q's temperature and humidity scheduling program and the weight of the items to be dried, the target temperature and humidity control device Q is locked to the reference temperature change curve S1 and reference humidity change curve S2 corresponding to the current scenario. Next, when the target temperature and humidity control device Q starts running and is in operation, the temperature and humidity data of the target temperature and humidity control device Q are monitored and collected in real time. The real-time temperature change curve S3 and the real-time humidity change curve S4 associated with the current operating state of the target temperature and humidity control device Q are constructed in chronological order. It should be noted that the real-time temperature change curve S3 and the real-time humidity change curve S4 are gradually improved over time.

[0030] Extract the real-time temperature change curve S3 and the real-time humidity change curve S4 of the target temperature and humidity control device Q, and plot them on the horizontal and vertical scales of the two-dimensional coordinate system XY1. Thus, the two-dimensional coordinate system XY1 includes four curves, namely the reference temperature change curve S1, the reference humidity change curve S2, the real-time temperature change curve S3, and the real-time humidity change curve S4. Using the total number of time points k preset by the operator as the dividing interval, starting from the 0 mark of the horizontal axis of the two-dimensional coordinate system XY1, the horizontal axis is divided to obtain j / k dividing interval segments. Determine the first segment, and draw straight lines perpendicular to the horizontal axis and parallel to the vertical axis through the 0 mark and the k mark on the segment, respectively. These are denoted as the first segment line L1 and the second segment line L2, and both the first segment line L1 and the second segment line L2 pass through S1, S2, S3, and S4. At this point, S1, S3, L1, and L2 will form one or more closed intervals. Calculate the area of ​​all closed intervals and denote it as the temperature difference characteristic TZH. Next, calculate the area of ​​one or more closed intervals formed by S2, S4, L1, and L2, and denote it as the humidity difference characteristic TZW; Then extract the temperature difference threshold H preset by the operator. yu Humidity difference threshold W yu ; The temperature difference characteristic TZH is compared with the temperature difference threshold H. yu Perform a comparison and simultaneously compare the humidity difference feature TZW with the humidity difference threshold W. yu Perform the comparison; If the comparison result is TZH≥H yu If the first temperature sensor malfunctions, an alarm will be issued and the temperature data acquisition method will be switched from the first temperature sensor to the second temperature sensor. Conversely, proceed with the following steps; If the comparison result is TZW≥W yu This indicates that the first humidity sensor has malfunctioned, and while outputting an alarm, the humidity data acquisition method is switched from the first humidity sensor to the second humidity sensor. Conversely, proceed with the following steps; If TZH < H yu And TZW < W yu The mixed drift characteristic SUM is calculated using SUM = TZH + TZW. Before adding TZH and TZW, a Taurus normalization operation is required to avoid inconsistencies in dimensions. And the associated anomaly feature SUM is compared with a preset mixed drift threshold SUM. yu Perform the comparison; if the comparison result is SUM ≥ SUM yuIf the first humidity sensor and the first temperature sensor are found to be experiencing mixed drift, a regular alarm will be output, and the operator will be notified to perform maintenance on the first humidity sensor and the first temperature sensor; otherwise, monitoring will continue. Using the above method, monitor all segmented intervals, identify abnormal changes, and output corresponding abnormal alarms.

[0031] All data in the formulas described above have been calculated with dimensions removed. Furthermore, any content not described in detail in this specification is existing technology known to those skilled in the art.

[0032] The above description is merely an example and illustration of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the invention or exceed the scope defined in the claims, all of which should fall within the protection scope of the present invention.

[0033] It should be stated that all user data collected in this application was collected with the user's consent and authorization. Furthermore, the uses of user data are legal and compliant, and the use and processing of user data comply with the relevant laws, regulations, and standards of the relevant regions.

Claims

1. A method for predicting and preventing abnormal temperature and humidity based on environmental adaptation, characterized in that, The method includes: Step 1: Monitor the target temperature and humidity control device based on the temperature and humidity detection module, collect the temperature and humidity data associated with the target temperature and humidity control device in the historical period, align them by time, and form a historical temperature and humidity data set. Step 2: Using the operating cycle as the interval, perform a partitioning operation on the historical temperature and humidity data set to determine the interval historical temperature data sequence and the interval historical humidity data sequence; Clustering is performed based on the weight of the items to be dried in each operating cycle to lock the secondary interval historical temperature data sequence and the secondary interval historical humidity data sequence of the items to be dried under different weights. Step 3: Construct the temperature change curve and humidity change curve associated with the secondary interval historical temperature data sequence and the secondary interval historical humidity data sequence of all items to be dried of the same weight. Perform a fitting operation to lock the reference temperature change curve and reference humidity change curve associated with the target temperature and humidity control equipment under the corresponding weight during the operating cycle. Step 4: Determine the weight of the items to be dried in the current target temperature and humidity control equipment, construct real-time temperature change curves and real-time humidity change curves, combine them with the corresponding weight's reference temperature change curves and reference humidity change curves to perform a verification operation, identify abnormal changes, and output an abnormal alarm.

2. The method according to claim 1, characterized in that, In step one, the specific method for constructing the historical temperature and humidity data set is as follows: Identify the target temperature and humidity control equipment, denoted as Q; Based on the temperature and humidity detection module, the temperature and humidity data of the target temperature and humidity control device Q at any time during any operating state within the historical period are obtained and constructed as historical temperature data sequences H1, H2, ..., Hj and historical humidity data sequences W1, W2, ..., Wj, respectively. Here, j is the total number of durations of the operating state, which is determined by the temperature and humidity scheduling program built into the temperature and humidity control device of the target temperature and humidity control device Q specification. Hi corresponds to Wi, i is the counting index, and 1≤i≤j. The historical period refers to a past time period, the specific duration of which is determined by the operator. The historical temperature data sequence H1, H2, ..., Hj and the historical humidity data sequence W1, W2, ..., Wj are combined and denoted as the associated temperature and humidity data combination D; Similarly, acquire temperature and humidity data of several temperature and humidity control devices of the same specifications as the target temperature and humidity control device Q when they are in operation during the historical period, and form a historical temperature data sequence, a historical humidity data sequence, and a combination of related temperature and humidity data. All the associated temperature and humidity data combinations constructed are summarized and denoted as the historical temperature and humidity data set {D1,D2,...,Dm}, where m is the total number of associated temperature and humidity data combinations.

3. The method according to claim 2, characterized in that, In step one, the temperature and humidity detection module includes a first temperature sensor and a second temperature sensor, as well as a first humidity sensor and a second humidity sensor. The first temperature sensor and the first humidity sensor are the main sensors. When the first temperature sensor and the first humidity sensor are abnormal, the system switches to the second temperature sensor and the second humidity sensor to perform data acquisition.

4. The method according to claim 3, characterized in that, In step two, the specific method for determining the interval historical temperature data sequence and the interval historical humidity data sequence is as follows: Obtain the historical temperature and humidity data set {D1,D2,...,Dm}; The statistical analysis includes all temperature and humidity scheduling programs for the target temperature and humidity control equipment of specification Q, denoted as CX1,CX2,...,CXu, where u is the total number of temperature and humidity scheduling programs. Get all associated temperature and humidity data combinations of all temperature and humidity control devices under any temperature and humidity scheduling program CXo during the operating cycle, and summarize them as the historical temperature and humidity data subset {Do1,Do2,...,Dor}, where r is the total number of associated temperature and humidity data combinations in the historical temperature and humidity data subset, 0≤r≤m, and o is the counting index, with a value range of 1≤o≤u; Separate the historical temperature data sequence and historical humidity data sequence corresponding to all associated temperature and humidity data combinations in the historical temperature and humidity data subset {Do1,Do2,...,Dor} and label them as interval historical temperature data sequence and interval historical humidity data sequence. Similarly, determine the subset of historical temperature and humidity data associated with the operating cycle corresponding to all temperature and humidity scheduling programs, as well as the interval historical temperature data sequence and interval historical humidity data sequence.

5. The method according to claim 4, characterized in that, In step two, the specific method for performing clustering based on the weight of the items to be dried within each operating cycle is as follows: Get the historical temperature and humidity data subset {Do1,Do2,...,Dor} associated with any temperature and humidity scheduler CXo; Simultaneously extract the weights of r associated temperature and humidity data combinations within the corresponding operating cycle, denoted as Go1, Go2, ..., Gor. Arrange Go1, Go2, ..., Gor in ascending order, and denote the weight sequence Go1', Go2', ..., Gor'. Using Go1' as the baseline, calculate the difference rate CY1 between Go1' and Go2', and compare it with the preset difference rate threshold CY. yu Comparison, if CY1≤CY yu Extract Go2', cluster it with Go1' into category F1, and continue to perform calibration operations with Go1' on subsequent weights; If there exists any weight, the difference rate between Goe' and Go1' is greater than CY yu Then, Goe' is used as the new baseline, and the unextracted weights are traversed. Clustering operations are performed to cluster into category F2, where e is the counting index and 1≤e≤r. This process continues until all weights in the weight sequence Go1',Go2',...,Gor' are clustered. The total number of clusters is counted and denoted as v. The historical temperature and humidity data subset {Do1,Do2,...,Dor} is clustered based on the weight of the items to be dried and their corresponding categories F1,F2,...,Fv. All interval historical temperature data sequences and interval historical humidity data sequences corresponding to any associated temperature and humidity data combinations in any category Fb are labeled as secondary interval historical temperature data sequences and secondary interval historical humidity data sequences, where 1≤b≤v.

6. The method according to claim 5, characterized in that, In step three, the specific method for locking the reference temperature change curve and reference humidity change curve associated with the target temperature and humidity control equipment under the corresponding weight during the operating cycle is as follows: Determine any associated temperature and humidity data combination Doe, which is classified into any category Fb, and its corresponding secondary interval historical temperature data sequence He1',He2',...,Hej' and secondary interval historical humidity data sequence We1',We2',...,Wej'; Construct a two-dimensional coordinate system XY1 with time as the horizontal axis, humidity as the positive half-axis, and temperature as the negative half-axis, and set the scale value of the negative half-axis to positive. The horizontal axis spans j time points. Align the historical temperature data sequence He1',He2',...,Hej' of the secondary interval with the historical humidity data sequence We1',We2',...,Wej' of the secondary interval with the horizontal axis scale, and plot them in the two-dimensional coordinate system XY1 in the form of data points to obtain two sets of data points. The data points are then fitted with curves and recorded as the temperature change curve and humidity change curve corresponding to the associated temperature and humidity data combination Doe, respectively. Similarly, determine the temperature change curves and humidity change curves corresponding to all associated temperature and humidity data combinations in category Fb; Fit all temperature change curves to obtain the reference temperature change curve S1. Similarly, by fitting all humidity change curves, the baseline humidity change curve S2 is obtained; As described above, determine the reference temperature change curve and reference humidity change curve for any weight scenario of any item to be dried under any temperature and humidity scheduling program.

7. The method according to claim 6, characterized in that, In step four, the specific method for determining the weight of the items to be dried in the current target temperature and humidity control equipment and constructing the real-time temperature change curve and the real-time humidity change curve is as follows: When the target temperature and humidity control device Q is in the ready-to-operate state, obtain the temperature and humidity scheduling program of the target temperature and humidity control device Q and the weight of the items to be dried. Based on the current target temperature and humidity control device Q's temperature and humidity scheduling program and the weight lock of the items to be dried, the corresponding reference temperature change curve S1 and reference humidity change curve S2 are used. When the target temperature and humidity control device Q is in operation, it monitors and collects temperature and humidity data in real time, and constructs real-time temperature change curve S3 and real-time humidity change curve S4 according to the time sequence.

8. The method according to claim 7, characterized in that, In step four, the specific method for identifying abnormal changes and outputting abnormal alarms is as follows: The real-time temperature change curve S3 and the real-time humidity change curve S4 are plotted in the two-dimensional coordinate system XY1 where the reference temperature change curve S1 and the reference humidity change curve S2 are located. Using the preset total number of time points k as the dividing interval, starting from the 0 mark of the horizontal axis of the two-dimensional coordinate system XY1, the horizontal axis is divided to determine the first dividing interval segment. Then, the straight lines perpendicular to the horizontal axis and parallel to the vertical axis are drawn through the 0 mark and the k mark on the dividing interval segment, respectively, and are recorded as the first dividing line L1 and the second dividing line L2. Calculate the area of ​​the closed interval formed by S1, S3, L1, and L2, and denote it as the temperature difference characteristic TZH; Calculate the area of ​​the closed interval formed by S2, S4, L1, and L2, and denote it as the humidity difference characteristic TZW; Extract the preset temperature difference threshold H yu Humidity difference threshold W yu ; The temperature difference feature TZH and humidity difference feature TZW are respectively compared with the temperature difference threshold H. yu Humidity difference threshold W yu Perform the comparison; If TZH≥H yu If the first temperature sensor is found to be malfunctioning, an alarm will be output and the system will switch to the second temperature sensor. If TZW≥W yu If the first humidity sensor is found to be malfunctioning, an alarm will be output and the system will switch to the second humidity sensor. If TZH < H yu And TZW < W yu The mixed drift characteristic SUM is calculated using SUM=TZH+TZW; The correlation anomaly feature SUM is compared with a preset mixed drift threshold SUM. yu Perform the comparison; If SUM≥SUM yu If the first humidity sensor and the first temperature sensor are found to be experiencing mixed drift, a standard alarm will be output, and the operator will be notified to perform maintenance. Conversely, continuous monitoring is required. Similarly, monitor all segmented intervals, identify abnormal changes, and output anomaly alerts.

9. A system for predicting and preventing abnormal temperature and humidity based on environmental adaptation, characterized in that, The system includes: The all-domain sensing and acquisition terminal monitors the target temperature and humidity control equipment based on the temperature and humidity detection module, and collects the temperature and humidity data associated with the target temperature and humidity control equipment in the historical period, aligns them by time, and forms a historical temperature and humidity data set. The spatiotemporal clustering decomposition end performs a partitioning operation on the historical temperature and humidity data set with the running cycle as the interval, and determines the interval historical temperature data sequence and interval historical humidity data sequence. Clustering is performed based on the weight of the items to be dried in each operating cycle to lock the secondary interval historical temperature data sequence and the secondary interval historical humidity data sequence of the items to be dried under different weights. In the baseline fitting modeling end, the temperature change curve and humidity change curve associated with the secondary interval historical temperature data sequence and the secondary interval historical humidity data sequence of all items to be dried of the same weight are constructed. The fitting operation is performed to lock the reference temperature change curve and reference humidity change curve associated with the target temperature and humidity control equipment under the corresponding weight during the operating cycle. The dynamic deviation verification terminal determines the weight of the items to be dried in the current target temperature and humidity control equipment, constructs real-time temperature change curves and real-time humidity change curves, and performs verification operations by combining the corresponding weight's reference temperature change curves and reference humidity change curves to identify abnormal changes and output abnormal alarms.

Images