An intelligent central water heater management method and system

Abstract

The application discloses an intelligent central water heater management method and system, and belongs to the technical field of intelligent buildings and energy-saving technology. According to abnormal operation data, the application sets the current operation strategy corresponding to the central water heater, thereby improving the efficiency of the central water heater in the use process, reducing energy consumption and saving cost. By constructing an isolated decision tree and obtaining abnormal operation data according to the isolated decision tree, the accuracy of the abnormal operation data acquisition result is improved, thereby further improving the efficiency.

Description

Technical Field

[0001] This invention relates to the fields of intelligent building technology and energy-saving technology, and in particular to an intelligent central water heater management method and system. Background Technology

[0002] A smart central water heater consists of one or more heat source systems that use electric heating to provide the hot water needed for a room.

[0003] Existing central water heaters operate according to user settings or system settings, referring to fixed parameters. That is, when a user needs to use the water, the user turns the central water heater on or off, or the building management equipment sets the operating time or operating temperature to meet the user's water needs.

[0004] However, the above management methods have limitations when user needs are diverse, resulting in situations where heating is not provided when users do not need water or when they do need water, leading to a poor user experience. Furthermore, the existing technology uses real-time heating to address these issues, which results in energy waste. Summary of the Invention

[0005] To address the problems of existing technologies, embodiments of the present invention provide an intelligent central water heater management method and system. The technical solution is as follows:

[0006] On the one hand, a method for managing an intelligent central water heater is provided, the method comprising:

[0007] Within the current time interval, acquire the historical operating data corresponding to the central water heater; the current time interval corresponds to the average ambient temperature.

[0008] Based on the historical operational data, construct an isolated decision tree;

[0009] The hierarchical distance between the leaf node and the root node corresponding to the historical running data in the isolated decision tree is adjusted to obtain the weighted distance;

[0010] A weighted anomaly score is calculated based on the weighted distance, and anomaly detection is performed based on the weighted anomaly score to determine abnormal operation data from the historical operation data;

[0011] Based on the abnormal operation data, set the current operation strategy for the central water heater.

[0012] Optionally, before acquiring the historical operating data corresponding to the central water heater within the current time interval, the method further includes:

[0013] Obtain the average value of the ambient temperature;

[0014] Based on the average value, set the corresponding time interval.

[0015] Optionally, constructing an isolated decision tree based on the historical operational data includes:

[0016] Calculate the first trend factor corresponding to the current time interval operation data based on the current time interval operation data in the historical operation data;

[0017] Calculate the second trend factor corresponding to the running data of the current time interval based on the fluctuation of the running data of the current time interval;

[0018] The isolated decision tree is constructed based on the first trend factor and the second trend factor.

[0019] Optionally, constructing the isolated decision tree based on the first trend factor and the second trend factor includes:

[0020] Calculate the average operating data of the central water heater within the current time interval;

[0021] Mapped running data is obtained based on the average running data within each running cycle and the average running data within the current time interval;

[0022] Based on the mapped running data, calculate the first entropy value corresponding to the current time interval and the second entropy value corresponding to the running cycle;

[0023] The isolated decision tree is constructed based on the first trend factor, the second trend factor, the first entropy value, and the second entropy value.

[0024] Optionally, constructing the isolated decision tree based on the first trend factor, the second trend factor, the first entropy value, and the second entropy value includes:

[0025] Based on the first entropy value and the second entropy value, determine the first selection weight of the mapping running data within the running cycle and the second selection weight within the time interval;

[0026] Based on the first selection weight, the second selection weight, the first trend factor, and the second trend factor, the data weights for different operating periods are determined;

[0027] The isolated decision tree is constructed based on the data weights.

[0028] Optionally, constructing the isolated decision tree based on the data weights includes:

[0029] Based on the data weights, the hierarchical distance between the leaf nodes and the root node corresponding to the historical running data in the isolated decision tree is adjusted to obtain a weighted distance, and a weighted anomaly score is calculated based on the weighted distance.

[0030] Anomaly detection is performed based on the weighted anomaly score, and abnormal operation data is determined from the historical operation data.

[0031] Optional,

[0032] The historical running data mean values ​​corresponding to each running cycle within the current time interval are sorted according to the time sequence to obtain the mean value sequence. The mean value sequence is then decomposed according to the STL time sequence decomposition method to obtain the slope of the trend term as the first trend factor.

[0033] The maximum absolute value of the difference between the historical running data of different running cycles within a day and the mean value of the data of each time interval is determined as the maximum difference, and the product of the maximum difference and the preset difference coefficient is calculated as the fluctuation coefficient.

[0034] The fluctuation coefficients of each time interval are sorted according to the time sequence to obtain the fluctuation sequence. The fluctuation sequence is then decomposed according to the STL time sequence decomposition method to obtain the slope of the trend term as the second trend factor.

[0035] Optional,

[0036] A trend influence coefficient is determined based on the first trend factor and the second trend factor, wherein the first trend factor and the trend influence coefficient are positively correlated, the second trend factor and the trend influence coefficient are positively correlated, and the value of the trend influence coefficient is a normalized value.

[0037] The normalized product of the trend influence coefficient, the first selection weight, and the second selection weight is calculated as the data weight.

[0038] Optional,

[0039] The product of the data weight and the hierarchical distance is calculated as the weighted distance;

[0040] Based on the outlier score calculation formula, the weighted distance is processed to calculate the outlier score, resulting in a weighted outlier score.

[0041] On the other hand, an intelligent central water heater management system is provided, the system comprising:

[0042] The acquisition device is used to acquire historical operating data of the central water heater within the current time interval; the current time interval corresponds to the average ambient temperature.

[0043] An analysis device is used to construct an isolated decision tree based on the historical operating data;

[0044] The analysis device is also used to adjust the hierarchical distance between the leaf nodes and the root node corresponding to the historical running data in the isolated decision tree to obtain a weighted distance;

[0045] The analysis device is also used to calculate a weighted anomaly score based on the weighted distance, perform anomaly detection based on the weighted anomaly score, and determine abnormal operation data from the historical operation data;

[0046] The strategy adjustment device is used to set the current operating strategy of the central water heater based on the abnormal operating data.

[0047] The beneficial effects of the technical solution provided by the embodiments of the present invention are as follows:

[0048] 1. By setting the corresponding operating strategy for the central water heater based on abnormal operating data, the efficiency of the central water heater during use is improved, energy consumption is reduced, and costs are saved;

[0049] 2. By constructing an isolated decision tree and obtaining abnormal operation data based on this isolated decision tree, the accuracy of abnormal operation data acquisition results is improved, thereby further improving efficiency. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a flowchart of the intelligent central water heater management method provided in an embodiment of the present invention;

[0052] Figure 2 This is a flowchart of the intelligent central water heater management method provided in an embodiment of the present invention;

[0053] Figure 3 This is a schematic diagram of an intelligent central water heater pipe system provided in an embodiment of the present invention. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0055] Reference Figure 1 As shown, a method for managing an intelligent central water heater is provided, the method including:

[0056] 101. Within the current time interval, obtain the historical operating data corresponding to the central water heater; the current time interval corresponds to the average ambient temperature;

[0057] 102. Construct an isolated decision tree based on historical operational data;

[0058] 103. Adjust the hierarchical distance between the leaf nodes and the root node corresponding to the historical running data in the isolated decision tree to obtain the weighted distance;

[0059] 104. Calculate the weighted anomaly score based on the weighted distance, perform anomaly detection based on the weighted anomaly score, and identify abnormal operating data from historical operating data;

[0060] 105. Based on abnormal operation data, set the current operation strategy corresponding to the central water heater.

[0061] In practical applications, this operating strategy can be specifically defined as follows:

[0062] Record the frequency and time of occurrence of abnormal operation data;

[0063] If the frequency of this abnormal operation data exceeds the preset value, then a fault detection will be performed on the central water heater;

[0064] If the test results indicate that the central water heater is not faulty, then at the next time the abnormal operation occurs, the central water heater will be operated according to the abnormal operation data.

[0065] If the test results indicate a malfunction in the central water heater, then the time interval between the two operating cycles with the longest time length within the time interval is selected as the repair time.

[0066] If the frequency of this abnormal operation data is less than or equal to the preset value, then a fault detection will be performed on the central water heater;

[0067] If the test results indicate that the central water heater is not faulty, the abnormal operating data is reported and marked. When the solution described in the embodiment of the present invention is executed in the next time interval, the abnormal operating data is ignored.

[0068] If the test results indicate a malfunction in the central water heater, then the time interval between the two operating cycles with the longest time length within the time interval is selected as the repair time.

[0069] If the frequency of this abnormal operation data is less than or equal to the preset value, then a fault detection will be performed on the central water heater.

[0070] If multiple instances of abnormal operating data occur during the use of a central water heater, assuming the water heater is functioning correctly, it indicates that the regular operating data does not meet current usage needs. The water heater will then be operated according to the abnormal data at the next occurrence, eliminating the need for users to adjust the operating data each time and thus improving user experience. If abnormal operating data occurs infrequently, and the water heater is functioning correctly, it can be attributed to user error or other improper use. Ignoring these abnormal operating data can prevent them from impacting the management of the central water heater.

[0071] In addition, by performing fault detection on the central water heater after abnormal data occurs, faults can be detected in a timely manner. The time interval between the two longest operating cycles is selected as the repair time, which ensures fault repair while avoiding inconvenience to users and further improves the user experience.

[0072] Optionally, before obtaining the historical operating data corresponding to the central water heater in step 101 within the current time interval, the method further includes:

[0073] 201. Obtain the average ambient temperature;

[0074] 202. Based on the average value, set the corresponding time interval to achieve the average value of the current time interval and the ambient temperature.

[0075] Specifically, the process can be as follows:

[0076] Multiple statistical periods are set, and the specific duration of the statistical period is not limited in this embodiment of the invention. In practical applications, the statistical period can be customized.

[0077] Obtain the average ambient temperature within the current statistical period;

[0078] If the average ambient temperature in the current statistical period and the average ambient temperature in the previous statistical period are both within the preset temperature range, and the absolute value of the difference between them is less than the preset value, then the current statistical period and the previous statistical period are set as the current time interval.

[0079] Continue performing the above operation for the next statistical period until the average ambient temperature within the statistical period is not at the preset temperature, or the absolute value of the difference between the average ambient temperatures of two statistical periods is greater than or equal to the preset value.

[0080] As is well known, the usage time and frequency of water heaters by users often correspond to the ambient temperature. It can be understood that users use water heaters more frequently and for longer periods when the ambient temperature is low, and vice versa. Therefore, by setting the current time interval to the average of the ambient temperature, the statistical time corresponds to the ambient temperature, which can avoid errors caused by insufficient consideration of the ambient temperature during the statistical process and further improve accuracy.

[0081] Optionally, step 102, which involves constructing an isolated decision tree based on historical operational data, includes:

[0082] 301. Based on the current time interval running data in the historical running data, calculate the first trend factor corresponding to the current time interval running data.

[0083] Specifically, the mean values ​​of historical running data corresponding to each running cycle within the current time interval are sorted according to time sequence to obtain a mean value sequence;

[0084] The mean series is decomposed using the STL time series decomposition method, and the slope of the trend term is used as the first trend factor.

[0085] The above operating cycle refers to the time between when the central water heater is turned on and when it is turned off.

[0086] By performing STL time series decomposition on the mean sequence, the corresponding trend term curve is obtained. Then, the slope of the trend term is used as the first trend factor. Since the trend term can characterize the overall trend of the corresponding time series, the slope of the trend term can characterize the changing trend of the running data of the running cycle within the corresponding time interval. That is, the first trend factor can characterize the changing trend of the running data of the running cycle within different time intervals, thereby obtaining abnormal running data more accurately.

[0087] 302. Based on the fluctuation of the running data at the current time interval, calculate the second trend factor corresponding to the running data at the current time interval.

[0088] Specifically, the maximum absolute value of the difference between the historical running data of different running cycles within the time interval and the mean value of the data within the time interval is determined as the maximum difference, and the product of the maximum difference and the preset difference coefficient is calculated as the fluctuation coefficient.

[0089] The volatility series is obtained by sorting the volatility coefficients of each time interval according to the time sequence. The volatility series is then decomposed according to the STL time series decomposition method to obtain the slope of the trend term as the second trend factor.

[0090] It should be noted that the time series decomposition (STL, Seasonal and Trend decomposition using Loess) method involved in calculating the trend factor in steps 301 and 302 is a time series decomposition method that uses robust local weighted regression as a smoothing method to decompose the time series into corresponding trend, seasonal and residual terms, which will not be elaborated further.

[0091] 303. Construct an isolated decision tree based on the first trend factor and the second trend factor.

[0092] In this embodiment of the invention, the mean sequence is decomposed using STL to obtain the corresponding trend term curve. Then, the slope of the trend term is used as the first trend factor. Since the trend term can characterize the overall trend of the corresponding time series, the slope of the trend term can characterize the changing trend of the mean of the running data corresponding to each running cycle at different time intervals. In other words, the first trend factor can characterize the changing trend of the mean of the running data corresponding to each running cycle at different time intervals, thereby obtaining abnormal running data more accurately.

[0093] The operational data described in this embodiment of the invention includes the start-up time, the stop-up time, and the heating temperature of the central water heater during operation. Correspondingly, the abnormal operational data includes the abnormal start-up time, the abnormal stop-up time, and the abnormal heating temperature.

[0094] Optional, refer to Figure 2 As shown, step 303, constructing an isolated decision tree based on the first trend factor and the second trend factor, includes:

[0095] 401. Calculate the average operating data of the central water heater within the current time interval;

[0096] 402. Based on the average value of the running data within each running cycle and the average value of the running data within the current time interval, obtain the mapped running data;

[0097] The mapping coefficient is obtained by using the sum of the average running data in each running cycle and the preset constant coefficient as the denominator and the average running data in the current time interval as the numerator.

[0098] The product of the historical running data and the mapping coefficient for the running cycle is calculated as the mapping running data for the corresponding running cycle.

[0099] 403. Based on the mapped running data, calculate the first entropy value corresponding to the current time interval and the second entropy value corresponding to the running cycle;

[0100] The first entropy value is determined based on the frequency distribution of different data within the time interval of the mapping operation data within the operation cycle, and the second entropy value is determined based on the frequency distribution of different data within the time interval of the mapping operation data.

[0101] 404. Construct an isolated decision tree based on the first trend factor, the second trend factor, the first entropy value, and the second entropy value.

[0102] Optionally, step 404, constructing an isolated decision tree based on the first trend factor, the second trend factor, the first entropy value, and the second entropy value, includes:

[0103] 501. Based on the first entropy value and the second entropy value, determine the first selection weight of the mapped running data within the running cycle and the second selection weight within the time interval;

[0104] 502. Determine the data weights for different operating periods based on the first selection weight, the second selection weight, the first trend factor, and the second trend factor;

[0105] 503. Construct an isolated decision tree based on data weights.

[0106] Optionally, step 503, constructing an isolated decision tree based on data weights, includes:

[0107] 601. Based on the data weights, adjust the hierarchical distance between the leaf nodes and the root node corresponding to the historical running data in the isolated decision tree to obtain the weighted distance, and calculate the weighted anomaly score based on the weighted distance;

[0108] 602. Detect outliers based on weighted outlier scores and identify abnormal operating data from historical operating data.

[0109] Optional,

[0110] The trend influence coefficient is determined based on the first trend factor and the second trend factor. The first trend factor and the trend influence coefficient are positively correlated, and the second trend factor and the trend influence coefficient are positively correlated. The value of the trend influence coefficient is a normalized value.

[0111] The normalized product of the trend influence coefficient, the first selection weight, and the second selection weight is used as the data weight.

[0112] Optional,

[0113] The weighted distance is calculated by multiplying the data weights and the hierarchical distances.

[0114] Based on the outlier score calculation formula, the weighted distance is processed to calculate the outlier score, resulting in a weighted outlier score.

[0115] This invention determines a first trend factor for historical operating data in each time interval by using the average data value for each time interval, and a second trend factor based on fluctuations. This effectively analyzes the trend factors based on data averages and fluctuations, ensuring accurate acquisition of historical operating data trends. By mapping historical operating data using the average data value for each time interval and the annual average data value, mapped water meter data is obtained. This eliminates objective differences in central water heater usage at different times, ensuring that the extraction of abnormal operating data avoids the influence of normal water heater usage. Since the first entropy value is determined based on the data frequency within the operating cycle, and based on the time interval... The frequency of data within a given period determines the second entropy value, enabling effective analysis of data within the corresponding operating cycle and time interval. This allows for the calculation of selection weights for the corresponding data based on the entropy value, resulting in the calculation of the first and second selection weights. These weights, along with the first and second trend factors, facilitate the determination of data weights. The data weights represent the impact weights of anomalies in the corresponding data. By combining the data weights with the hierarchical distance, a weighted distance is obtained. Based on this weighted distance, an adaptive weighted anomaly score is derived. This allows the weighted anomaly score to effectively characterize anomalies in the corresponding historical operating data, thereby improving the detection accuracy and precision of anomaly detection.

[0116] Reference Figure 3 As shown, an intelligent central water heater management system is provided, the system including:

[0117] The acquisition device is used to acquire historical operating data of the central water heater within the current time interval; the current time interval corresponds to the average ambient temperature.

[0118] An analysis device is used to construct isolated decision trees based on historical operational data;

[0119] The analysis device is also used to adjust the hierarchical distance between the leaf nodes and the root node corresponding to the historical running data in the isolated decision tree to obtain the weighted distance;

[0120] The analysis device is also used to calculate a weighted anomaly score based on the weighted distance, detect anomalies based on the weighted anomaly score, and identify abnormal operating data from historical operating data;

[0121] The strategy adjustment device is used to set the current operating strategy of the central water heater based on abnormal operating data.

[0122] Optionally, the acquisition device is specifically used for:

[0123] Obtain the average ambient temperature;

[0124] Set the corresponding time interval based on the average value.

[0125] Optionally, the analytical device is specifically used for:

[0126] Calculate the first trend factor corresponding to the current time interval running data based on the historical running data;

[0127] Calculate the second trend factor corresponding to the running data at the current time interval based on the fluctuation of the running data at the current time interval;

[0128] Construct an isolated decision tree based on the first and second trend factors.

[0129] Optionally, the analytical device is specifically used for:

[0130] Calculate the average operating data of the central water heater within the current time interval;

[0131] Mapped running data is obtained based on the average running data within each running cycle and the average running data within the current time interval;

[0132] Based on the mapped running data, calculate the first entropy value corresponding to the current time interval and the second entropy value corresponding to the running cycle;

[0133] Construct an isolated decision tree based on the first trend factor, the second trend factor, the first entropy value, and the second entropy value.

[0134] Optionally, the analytical device is specifically used for:

[0135] Based on the first entropy value and the second entropy value, determine the first selection weight of the mapping running data within the running cycle and the second selection weight within the time interval;

[0136] The data weights for different operating periods are determined based on the first selection weight, the second selection weight, the first trend factor, and the second trend factor.

[0137] Construct an isolated decision tree based on data weights.

[0138] Optionally, the analytical device is specifically used for:

[0139] Based on the data weights, the hierarchical distance between the leaf nodes and the root node corresponding to the historical running data in the isolated decision tree is adjusted to obtain the weighted distance, and the weighted anomaly score is calculated based on the weighted distance.

[0140] Anomaly detection is performed based on weighted anomaly scores, and abnormal operational data is identified from historical operational data.

[0141] Optionally, the analytical device is specifically used for:

[0142] The historical running data mean values ​​corresponding to each running cycle within the current time interval are sorted according to the time sequence to obtain the mean value sequence. The mean value sequence is then decomposed according to the STL time sequence decomposition method to obtain the slope of the trend term as the first trend factor.

[0143] The maximum absolute value of the difference between the historical running data of different running cycles within a day and the mean of the data for each time interval is determined as the maximum difference, and the product of the maximum difference and the preset difference coefficient is calculated as the fluctuation coefficient.

[0144] The volatility series is obtained by sorting the volatility coefficients of each time interval according to the time sequence. The volatility series is then decomposed according to the STL time series decomposition method to obtain the slope of the trend term as the second trend factor.

[0145] Optionally, the analytical device is specifically used for:

[0146] The trend influence coefficient is determined based on the first trend factor and the second trend factor. The first trend factor and the trend influence coefficient are positively correlated, and the second trend factor and the trend influence coefficient are positively correlated. The value of the trend influence coefficient is a normalized value.

[0147] The normalized product of the trend influence coefficient, the first selection weight, and the second selection weight is used as the data weight.

[0148] Optional,

[0149] The weighted distance is calculated by multiplying the data weights and the hierarchical distances.

[0150] Based on the outlier score calculation formula, the weighted distance is processed to calculate the outlier score, resulting in a weighted outlier score.

[0151] It should be noted that, in the context of smart buildings or smart structures, the intelligent central water heater management system described in the embodiments of the present invention can specifically involve setting corresponding detection points on the user side or on the user's corresponding home appliances; for example, setting a corresponding detection point on the user's hot water inlet to detect the temperature of the hot water used by the user in real time; and configuring a strategy adjustment device, an analysis device, and an acquisition device on the building management equipment to realize the technical solution provided in the embodiments of the present invention.

[0152] In practical applications, considering the heat loss between the central water heater and the user side, the central water heater can be configured according to the floors or according to users on a certain number of floors, so that multiple central water heaters can be configured in the same building. While ensuring use, this further improves the management precision of the central water heater, ensures that user needs are met, and improves the user experience.

[0153] Central water heaters can be configured according to the number of floors or according to users on a certain number of floors. During this process, heat loss can also be calculated by comparing the temperature of the hot water used by the users with the heating temperature of the central water heater in real time.

[0154] The technical solution provided by the embodiments of the present invention avoids conceptual shifts in the judgment model during the learning process, which leads to reduced accuracy in data anomaly detection and insufficient accuracy in anomaly detection when the judgment model is trained with random data.

[0155] It should be noted that the intelligent central water heater management system provided in the above embodiments is only illustrated by the division of the above functional modules when executing the intelligent central water heater management method. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the system can be divided into different functional modules to complete all or part of the functions described above. In addition, the intelligent central water heater management method and system embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0156] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0157] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for managing an intelligent central water heater, characterized in that, The method includes: Within the current time interval, acquire the historical operating data corresponding to the central water heater; the current time interval corresponds to the average ambient temperature. Based on the historical operational data, construct an isolated decision tree; The hierarchical distance between the leaf node and the root node corresponding to the historical running data in the isolated decision tree is adjusted to obtain the weighted distance; A weighted anomaly score is calculated based on the weighted distance, and anomaly detection is performed based on the weighted anomaly score to determine abnormal operation data from the historical operation data; Based on the abnormal operation data, a new operating strategy is set for the central water heater, wherein the operating strategy includes: Record the frequency and time of occurrence of the abnormal operation data; If the frequency of the abnormal operation data is greater than a preset value, then the central water heater will be tested for faults. If the central water heater is not faulty, then at the next time an abnormality occurs, the central water heater will be operated according to the abnormal operation data. If the central water heater malfunctions, the time interval between all adjacent operating cycles within the time interval is obtained, and the time interval between the two operating cycles with the longest time length is selected as the repair time. Before acquiring the historical operating data corresponding to the central water heater within the current time interval, the method further includes: Obtain the average value of the ambient temperature; Based on the average value, a corresponding time interval is set, including: Set multiple statistical periods; Obtain the average value of the ambient temperature within the current statistical period; If the average ambient temperature in the current statistical period and the average ambient temperature in the previous statistical period are both within a preset temperature range, and the absolute value of the difference between them is less than a preset value, then the current statistical period and the previous statistical period are set as the current time interval. Continue performing the above operation for the next statistical period until the average ambient temperature within the statistical period is not at the preset temperature, or the absolute value of the difference between the average ambient temperatures of two statistical periods is greater than or equal to the preset value.

2. The method according to claim 1, characterized in that, The construction of the isolated decision tree based on the historical operational data includes: Calculate the first trend factor corresponding to the current time interval operation data based on the current time interval operation data in the historical operation data; Calculate the second trend factor corresponding to the running data of the current time interval based on the fluctuation of the running data of the current time interval; The isolated decision tree is constructed based on the first trend factor and the second trend factor.

3. The method according to claim 2, characterized in that, The step of constructing the isolated decision tree based on the first trend factor and the second trend factor includes: Calculate the average operating data of the central water heater within the current time interval; Mapped running data is obtained based on the average running data within each running cycle and the average running data within the current time interval; Based on the mapped running data, calculate the first entropy value corresponding to the current time interval and the second entropy value corresponding to the running cycle; The isolated decision tree is constructed based on the first trend factor, the second trend factor, the first entropy value, and the second entropy value.

4. The method according to claim 3, characterized in that, The step of constructing the isolated decision tree based on the first trend factor, the second trend factor, the first entropy value, and the second entropy value includes: Based on the first entropy value and the second entropy value, determine the first selection weight of the mapping running data within the running cycle and the second selection weight within the time interval; Based on the first selection weight, the second selection weight, the first trend factor, and the second trend factor, the data weights for different operating periods are determined; The isolated decision tree is constructed based on the data weights.

5. The method according to claim 4, characterized in that, The step of constructing the isolated decision tree based on the data weights includes: Based on the data weights, the hierarchical distance between the leaf nodes and the root node corresponding to the historical running data in the isolated decision tree is adjusted to obtain a weighted distance, and a weighted anomaly score is calculated based on the weighted distance. Anomaly detection is performed based on the weighted anomaly score, and abnormal operation data is determined from the historical operation data.

6. The method according to claim 5, characterized in that, The historical running data mean values ​​corresponding to each running cycle within the current time interval are sorted according to the time sequence to obtain the mean value sequence. The mean value sequence is then decomposed according to the STL time sequence decomposition method to obtain the slope of the trend term as the first trend factor. The maximum absolute value of the difference between the historical running data of different running cycles within a day and the mean value of the data of each time interval is determined as the maximum difference, and the product of the maximum difference and the preset difference coefficient is calculated as the fluctuation coefficient. The fluctuation coefficients of each time interval are sorted according to the time sequence to obtain the fluctuation sequence. The fluctuation sequence is then decomposed according to the STL time sequence decomposition method to obtain the slope of the trend term as the second trend factor.

7. The method according to claim 6, characterized in that, A trend influence coefficient is determined based on the first trend factor and the second trend factor, wherein the first trend factor and the trend influence coefficient are positively correlated, the second trend factor and the trend influence coefficient are positively correlated, and the value of the trend influence coefficient is a normalized value. The normalized product of the trend influence coefficient, the first selection weight, and the second selection weight is calculated as the data weight.

8. The method according to claim 7, characterized in that, The product of the data weight and the hierarchical distance is calculated as the weighted distance; Based on the outlier score calculation formula, the weighted distance is processed to calculate the outlier score, resulting in a weighted outlier score.

9. An intelligent central water heater management system, characterized in that, The system includes: The acquisition device is used to acquire historical operating data of the central water heater within the current time interval; the current time interval corresponds to the average ambient temperature. An analysis device is used to construct an isolated decision tree based on the historical operating data; The analysis device is also used to adjust the hierarchical distance between the leaf nodes and the root node corresponding to the historical running data in the isolated decision tree to obtain a weighted distance; The analysis device is also used to calculate a weighted anomaly score based on the weighted distance, perform anomaly detection based on the weighted anomaly score, and determine abnormal operation data from the historical operation data; A strategy adjustment device is used to set the current operating strategy of the central water heater based on the abnormal operating data, wherein the operating strategy includes: Record the frequency and time of occurrence of the abnormal operation data; If the frequency of the abnormal operation data is greater than a preset value, then the central water heater will be tested for faults. If the central water heater is not faulty, then at the next time an abnormality occurs, the central water heater will be operated according to the abnormal operation data. If the central water heater malfunctions, the time interval between all adjacent operating cycles within the time interval is obtained, and the time interval between the two operating cycles with the longest time length is selected as the repair time. The acquisition device is specifically used for: Obtain the average value of the ambient temperature; Based on the average value, a corresponding time interval is set, including: Set multiple statistical periods; Obtain the average value of the ambient temperature within the current statistical period; If the average ambient temperature in the current statistical period and the average ambient temperature in the previous statistical period are both within a preset temperature range, and the absolute value of the difference between them is less than a preset value, then the current statistical period and the previous statistical period are set as the current time interval. Continue performing the above operation for the next statistical period until the average ambient temperature within the statistical period is not at the preset temperature, or the absolute value of the difference between the average ambient temperatures of two statistical periods is greater than or equal to the preset value.

Images