Heat supply anomaly identification method and device based on instantaneous flow per unit area deviation

Abstract

The application discloses a heat supply anomaly identification method and device based on unit area instantaneous flow deviation, and relates to the technical field of heat supply system operation and maintenance. The method comprises the following steps: collecting the real-time heat supply flow, the rated design heat supply flow and the corresponding heat supply building area of the target heat supply facility of a user in a heat supply system; calculating the unit area instantaneous flow relative deviation according to the three data, so as to eliminate the interference of the heat supply area difference on the flow determination; starting a time window with a preset time length, and cumulatively counting the unit area instantaneous flow relative deviations in the window, so as to filter the influence of instantaneous fluctuation on the determination result; when the deviation after the cumulative counting exceeds a preset abnormal threshold, the unit area instantaneous flow relative deviation of a neighboring user on the same floor is obtained; finally, according to the comparison result of the deviation of the neighboring user and the abnormal threshold, it is identified whether the current anomaly belongs to a single-household local anomaly or a building-level global hydraulic imbalance, and the corresponding abnormal level is output.

Description

Technical Field

[0001] This application relates to the field of heating system operation and maintenance technology, and in particular to a method and device for identifying heating anomalies based on instantaneous flow deviation per unit area. Background Technology

[0002] In the field of operation and maintenance technology for centralized heating systems, heating companies or property management departments sometimes adopt technical solutions based on heating flow data to identify abnormal states in order to ensure heating quality for users, reduce energy consumption, or detect equipment failures in a timely manner. For example, they may monitor whether the instantaneous flow of a user or building is lower than the normal range to determine whether there are faults such as pipe blockage or valve jamming. In existing technologies, the absolute value of the flow is generally used directly as the basis for judgment. For example, a fixed flow threshold is set, and when the measured flow is lower than the threshold, it is judged as abnormal. However, due to the large differences in heating area among different users, the baseline flow of large-area users is naturally different from that of small-area users. The absolute value index cannot achieve normalized comparison across users, which can easily lead to misjudging normal low flow of large-area users as abnormal, or missing the real blockage of small-area users. At the same time, most existing solutions only rely on single-point instantaneous data for judgment, ignoring short-term interference conditions such as temporary water release by users and instantaneous fluctuations in water pressure, resulting in frequent false alarms. Summary of the Invention

[0003] This specification provides a method and apparatus for identifying heating anomalies based on instantaneous flow deviation per unit area, in order to solve at least one of the technical problems mentioned above.

[0004] To solve the above-mentioned technical problems, the embodiments in this specification are implemented as follows: According to a first aspect of the embodiments of this specification, a method for identifying heating anomalies based on instantaneous flow rate deviation per unit area is provided, comprising: Collect the real-time heating flow rate, rated design heating flow rate, and corresponding heating building area of ​​the target heating facilities of users in the heating system. Based on the real-time heating flow rate, the rated design heating flow rate, and the heating building area, calculate the relative deviation of the instantaneous flow rate per unit area; A preset time window is started, and the relative deviations of instantaneous flow rates per unit area within the time window are accumulated and statistically analyzed to filter out instantaneous fluctuations; When the absolute value of the cumulative deviation exceeds the preset abnormal threshold, the relative deviation of the instantaneous flow per unit area of ​​adjacent users on the same floor is obtained. Based on the comparison between the relative deviation of instantaneous flow per unit area of ​​adjacent users and the preset abnormal threshold, abnormal states are identified and corresponding abnormal levels are output. The formula for calculating the relative deviation of instantaneous flow rate per unit area is as follows: In the formula, The relative deviation of instantaneous flow rate per unit area. Real-time heat flow rate for the user's target heating facility. The rated design heating flow rate for the user's target heating facility.

[0005] In some optional implementations, the step of calculating the relative deviation of instantaneous flow rate per unit area based on the real-time heating flow rate, the rated design heating flow rate, and the heating building area specifically includes: Calculate the real-time flow rate per unit area by dividing the real-time heating flow rate by the heating building area; The design rated flow rate per unit area is calculated by dividing the rated design heating flow rate by the heating building area. The absolute deviation of the flow rate per unit area is calculated as the real-time flow rate per unit area minus the design rated flow rate per unit area. Divide the absolute deviation of the flow rate per unit area by the design rated flow rate per unit area and multiply by 100% to obtain the relative deviation of the instantaneous flow rate per unit area. The step of accumulating and statistically analyzing the relative deviations of instantaneous flow rate per unit area within a preset time window includes: setting a sliding time window of fixed duration; continuously collecting a preset number of relative deviations of instantaneous flow rate per unit area within the time window at a preset sampling period; calculating the arithmetic mean of the preset number of relative deviations of instantaneous flow rate per unit area as the accumulated deviation; wherein, the formula for calculating the accumulated deviation is: In the formula, The cumulative average normalized deviation over the time window. For the first in the window Instantaneous normalized flow deviation for each sampling period This represents the number of valid sampling points within the window. Take a preset positive integer; The step of accumulating and statistically analyzing the relative deviations of instantaneous flow rates per unit area within a time window with a preset duration includes: maintaining the time window using a mechanism of sliding and updating each sampling period; removing the oldest historical data within the window for each new sampled data, so that the window always maintains the latest preset number of continuous sampled data.

[0006] In some optional implementations, the step of activating a preset time window and cumulatively calculating the relative deviations of instantaneous flow rates per unit area within the time window further includes: Calculate the stability index of the relative deviation sequence of instantaneous flow rate per unit area within the time window. and trend change rate : In the formula, Characterizing the instantaneous fluctuation intensity of the deviation sequence, The rate of absolute value deterioration of the biased sequence; When the stability index The trend change rate is less than the preset stability threshold and the trend change rate is less than the preset stability threshold. When the current anomaly exceeds the preset deterioration threshold, it is determined to be a gradual traffic anomaly fault, and the length of the time window is dynamically reduced to a preset proportion of the original length to accelerate the anomaly response. When the stability index When the current anomaly is greater than or equal to the preset stability threshold, it is determined to be a transient disturbance or a sudden fault, and the original length of the time window is maintained.

[0007] In some optional implementations, in the step of obtaining the relative deviation of instantaneous flow per unit area of ​​adjacent users on the same floor, the adjacent users are a preset number of users who are pre-selected, are on the same floor as the target user, are on the same heating riser, and whose heating area deviation does not exceed a preset ratio. The step of obtaining the relative deviation of instantaneous flow per unit area of ​​adjacent users on the same floor when the absolute value of the cumulative statistical deviation exceeds the preset abnormal threshold is as follows: when the absolute value of the cumulative statistical deviation of a preset number of consecutive sampling periods exceeds the first preset threshold, cross-validation of neighboring households is triggered to obtain the relative deviation of instantaneous flow per unit area of ​​adjacent users on the same floor.

[0008] In some optional implementations, the step of identifying the abnormal state and outputting the corresponding abnormal level based on the comparison result of the relative deviation of the instantaneous flow per unit area of ​​the adjacent users with the preset abnormal threshold includes: If the absolute value of the cumulative statistical deviation of the target user is greater than or equal to the first preset threshold, and the absolute value of the cumulative statistical deviation of its adjacent users in the same layer is less than or equal to the second preset threshold, then it is determined to be a single-household local anomaly, and the first anomaly level is output. If the absolute value of the cumulative deviation of the target user is greater than or equal to the first preset threshold, and the absolute value of the cumulative deviation of the adjacent users on the same floor exceeding a preset proportion is greater than or equal to the third preset threshold, then it is determined to be a building-level global hydraulic imbalance, and the second abnormal level is output.

[0009] In some alternative implementations, the following steps are also included: Construct an anomaly severity index The calculation formula is as follows: In the formula, The deviation is calculated cumulatively for the target users. The first preset threshold, The first nonlinear weighting coefficient, The number of adjacent users, For the first The deviation after cumulative statistics of adjacent users The second preset threshold or the third preset threshold, The second nonlinear weighting coefficient, The standard deviation of the cumulative deviations of all adjacent users. This is the deviation dispersion adjustment coefficient; According to the severity index of the anomaly The preset value range in which the first or second abnormality level is located is used to refine the first or second abnormality level into multiple sub-levels, and corresponding sub-level warning signals are output.

[0010] In some alternative implementations, the following steps are also included: Calculate the spatial dispersion index of the deviation between the target user and all its neighboring users. and homogeneity coefficient : In the formula, The number of adjacent users, For the first The deviation after cumulative statistics of adjacent users The mean of the absolute values ​​of the deviations between adjacent users. Use extremely small positive numbers to prevent division by zero errors; When the homogeneity coefficient Within a preset global imbalance range and the spatial dispersion index When the deviation is less than the preset upper limit of dispersion, it is determined that the deviation of adjacent user groups is in the same direction, the fault type is building-level global hydraulic imbalance, and the second abnormal level is upgraded to the first-level warning priority. When the homogeneity coefficient The spatial dispersion index is greater than or equal to the first homogeneity threshold and the spatial dispersion index When the dispersion is less than the preset upper limit, the target user is determined to be an isolated anomaly, the fault type is a single-household local anomaly, and the first anomaly level is maintained. When the homogeneity coefficient The spatial dispersion index is located in the preset middle range. When the deviation exceeds the preset lower limit of dispersion, it is determined that the deviations of adjacent user groups are independent of each other, and the fault type is multiple isolated local anomalies superimposed, each maintaining its own first anomaly level output.

[0011] In some optional implementations, the step of collecting the supply and return water temperature difference of the user's target heating facility or building as a multi-dimensional feature is also included to assist in verifying the abnormal state.

[0012] In some optional implementations, the step of identifying the abnormal state and outputting the corresponding abnormal level based on the comparison result of the relative deviation of the instantaneous flow per unit area of ​​the adjacent users with the preset abnormal threshold further includes: Obtain the relative deviation sequence of instantaneous flow per unit area for the target user within the time window. and the corresponding supply and return water temperature difference sequence ; Calculate the time coupling coefficient between the deviation sequence and the supply and return water temperature difference sequence. : In the formula, The arithmetic mean of the supply and return water temperature difference within the time window; Based on the absolute value of the timing coupling coefficient The first preset threshold, the second preset threshold, and the third preset threshold are dynamically adjusted to obtain the adjusted dynamic threshold. The first preset threshold, the second preset threshold, and the third preset threshold are replaced with the corrected dynamic threshold, and the anomaly determination and level output are re-executed.

[0013] According to a second aspect of the embodiments of this specification, a heating anomaly identification device based on instantaneous flow rate deviation per unit area is provided, comprising: The data acquisition module is used to collect the real-time heating flow rate, rated design heating flow rate, and corresponding heating building area of ​​the target heating facilities of users in the heating system. The normalized deviation calculation module is used to calculate the relative deviation of instantaneous flow per unit area based on the real-time heating flow, rated design heating flow, and heating building area. The sliding window cumulative filtering module is used to start a time window of preset duration and accumulate and statistically analyze the relative deviations of instantaneous flow rates per unit area within the time window in order to filter out instantaneous fluctuations. The neighboring household data acquisition module is used to acquire the relative deviation of instantaneous flow per unit area of ​​adjacent users on the same floor when the absolute value of the cumulative statistical deviation exceeds the preset abnormal threshold. The anomaly detection and graded output module is used to identify anomalies and output corresponding anomaly levels based on the comparison results between the relative deviation of the instantaneous flow per unit area of ​​the adjacent users and the preset anomaly threshold. The formula for calculating the relative deviation of instantaneous flow rate per unit area is as follows: In the formula, The relative deviation of instantaneous flow rate per unit area. Real-time heat flow rate for the user's target heating facility. The rated design heating flow rate for the user's target heating facility.

[0014] One embodiment of this specification can achieve at least the following beneficial effects: In this technical solution, after collecting the user's real-time heating flow rate, instead of directly using the absolute flow rate value as the basis for anomaly judgment, the relative deviation of the instantaneous flow rate per unit area is first calculated using the rated design flow rate and the heating building area, thereby eliminating numerical differences caused by different heating areas. Subsequently, a preset time window is activated, and multiple deviation values ​​within the window are accumulated and statistically analyzed to filter out transient interferences such as temporary water releases or water pressure fluctuations by users. Based on this, if the accumulated deviation exceeds the anomaly threshold, similar deviation data of adjacent users on the same floor are further obtained, and the anomaly of the target user is cross-validated using the deviation status of neighboring households. Since the neighboring households and the target user are on the same floor and on the same heating riser, their pipe network conditions are basically the same, and comparison can effectively distinguish between local faults in a single household and overall hydraulic imbalance in the building. In this way, the entire identification process no longer relies on the original absolute flow rate value or the instantaneous value at a single sampling moment, but undergoes multiple verifications through normalization, cumulative filtering, and neighborhood comparison, thereby ensuring the accuracy and reliability of anomaly judgment to a certain extent. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of a heating anomaly identification method based on instantaneous flow rate deviation per unit area, provided in the embodiments of this specification. Figure 2 This is a schematic diagram of the screening rules for adjacent users on the same floor and on the same riser in the heating anomaly identification method based on instantaneous flow deviation per unit area provided in the embodiments of this specification. Figure 3 This is a comparison table of the distribution of instantaneous flow deviation per unit area for each user in the scenario of global hydraulic imbalance at the building level, in the heating anomaly identification method based on instantaneous flow deviation per unit area provided in the embodiments of this specification; Figure 4 This is a schematic diagram of the gradual flow anomaly fault propagation process in the heating anomaly identification method based on instantaneous flow deviation per unit area provided in the embodiments of this specification; Figure 5 This is an illustration of the color-coded anomaly level in the heating anomaly identification method based on instantaneous flow deviation per unit area provided in the embodiments of this specification; Figure 6 The embodiments provided in this specification correspond to Figure 1 A schematic diagram of a heating anomaly identification device based on instantaneous flow deviation per unit area. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of one or more embodiments of this specification clearer, the technical solutions of one or more embodiments of this specification will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of them. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of one or more embodiments of this specification.

[0018] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another.

[0019] The technical solution of this application will be described in detail below with reference to the accompanying drawings, such as... Figure 1 As shown, the method may include: Step 102: Collect the real-time heating flow rate, rated design heating flow rate, and corresponding heating building area of ​​the target heating facilities of users in the heating system.

[0020] In the embodiments of this specification, a target heating facility can refer to a heating terminal device in a centralized heating system corresponding to a specific user. For the user, this target heating facility can provide heating services to the building space, such as inlet pipes, radiators, or underfloor heating coils, which are terminal components that perform heat release functions. A heating company or property management company operates target heating facilities covering a large number of users, thereby ensuring the winter heating needs of each user. In practical applications, from a data acquisition perspective, front-end metering devices such as flow meters or heat meters can be configured to transmit the real-time operating data collected from each user's end back to the back-end data processing platform via network connection, thereby enabling the monitoring of the operating status of each dispersed target heating facility. Real-time heating flow rate can refer to the volumetric flow rate value read by the flow meter installed on the user's inlet pipe at the current sampling time, with the unit being m³ / h, reflecting the actual circulation volume of the heat medium through the user's heating facility per unit time. Rated design heating flow rate refers to the baseline flow rate calculated during the planning and design phase of the user's heating facilities based on its heating area, building envelope thermal parameters, and design supply and return water temperatures. Once determined, this value is stored as an inherent design parameter for the user in the backend database and does not change with actual operating conditions. Heated building area refers to the actual heating area of ​​the building space served by the user's heating facilities, which can be derived from architectural design drawings or stipulated in the heating contract.

[0021] In practical applications, ultrasonic flow meters or mechanical heat meters deployed on the inlet pipes of each user can continuously read instantaneous flow values ​​at a sampling period of minutes. Then, the real-time flow data is reported to the data aggregation node through a data acquisition device or wired / wireless communication module. At the same time, the background database retrieves the rated design flow and heating area data pre-entered by the user, and aligns the three according to the user code and timestamp, thereby providing complete basic data input for subsequent normalized deviation calculation.

[0022] Step 104: Calculate the relative deviation of instantaneous flow rate per unit area based on the real-time heating flow rate, rated design heating flow rate, and heating building area. The formula for calculating the relative deviation of instantaneous flow rate per unit area is: In the formula, The relative deviation of instantaneous flow rate per unit area. Real-time heat flow rate for the user's target heating facility. The rated design heating flow rate for the user's target heating facility.

[0023] In this step, the real-time heating flow rate collected from the user is first divided by the corresponding heating building area to obtain the real-time flow rate per unit area. Simultaneously, the user's rated design heating flow rate is divided by the same heating building area to obtain the rated design flow rate per unit area. These two values ​​reflect the flow load levels corresponding to a unit heating area under actual operating conditions and design standard operating conditions, respectively. Then, the difference between the real-time flow rate per unit area and the rated design flow rate per unit area is calculated to obtain the absolute deviation of the flow rate per unit area. Finally, this absolute deviation is divided by the rated design flow rate per unit area and multiplied by 100% to obtain the relative deviation of the instantaneous flow rate per unit area. This calculation can be uniformly expressed as a formula. ,in The flow rate value collected in real time for the target heating facility. This refers to the standard flow rate value determined during the heating design phase. The relative deviation of the instantaneous flow rate per unit area calculated in this step... The value no longer depends on the user's specific area, but is directly determined by the ratio of real-time traffic to designed traffic. Therefore, whether the user is a large apartment of 100㎡ or a small apartment of 50㎡, as long as the relative deviation is the same, it represents the same degree of anomaly severity. This can provide a unified and comparable data basis for subsequent time window filtering and cross-validation of neighboring households.

[0024] Step 106: Start a time window of preset duration, and accumulate and statistically analyze the relative deviations of instantaneous flow rates per unit area within the time window to filter out instantaneous fluctuations.

[0025] This step initiates a sliding time window of a preset duration (e.g., 5 minutes), continuously collecting relative deviation data of instantaneous flow rate per unit area at preset sampling intervals (e.g., 1 minute). The arithmetic mean of all sampling points within the window is calculated as the cumulative statistical deviation. This mechanism effectively distinguishes between transient disturbances lasting only 1-3 minutes (such as temporary water release by users or instantaneous fluctuations in water pressure) and persistent real anomalies (such as pipe blockage or valve stagnation) through the mean filtering effect of the time window. In the former case, individual spikes within the window are averaged and insufficient to trigger the threshold, while in the latter case, the deviation remains consistently high over multiple sampling intervals, causing the window mean to stably exceed the threshold. This avoids false alarms caused by single-point transient fluctuations at the algorithm's root. Simultaneously, the window employs a sliding update mechanism across sampling intervals, removing the oldest data with each new sample to ensure the window always maintains the latest continuous data, balancing filtering stability and timely response.

[0026] Step 108: When the absolute value of the accumulated deviation exceeds the preset abnormal threshold, obtain the relative deviation of instantaneous flow per unit area of ​​adjacent users on the same floor.

[0027] This step triggers the acquisition of neighboring household data only when the absolute value of the accumulated deviation over the time window reaches or exceeds a preset anomaly threshold. The reason for using the absolute value of the deviation as the criterion is that heating anomalies can manifest as either insufficient flow (negative deviation, e.g., -35%) or excessive flow (positive deviation, e.g., +35%). Taking the absolute value unifies both types of deviations into the judgment range, avoiding the omission of either type of anomaly due to different sign directions. This trigger condition ensures that the cross-validation process with neighboring users is only initiated when the target user's deviation is confirmed as a persistent and significant anomaly, thus avoiding increased computational burden and communication overhead due to frequent neighboring household data queries. When the trigger condition is met, the system acquires the instantaneous relative flow deviation per unit area of ​​a preset number of neighboring users located on the same floor and on the same heating riser as the target user, with heating area deviations not exceeding a preset proportion. This data has undergone the same normalization calculation and time window filtering, ensuring strict consistency with the target user data in terms of dimension, time reference, and physical meaning, thus providing a reliable basis for subsequent anomaly type determination through horizontal comparison.

[0028] Step 110: Based on the comparison result between the relative deviation of the instantaneous flow per unit area of ​​the adjacent users and the preset abnormal threshold, identify the abnormal state and output the corresponding abnormal level.

[0029] The purpose of this step is to compare the deviation of the target user with the deviation of adjacent user groups to accurately distinguish the root cause of the anomaly, i.e., whether it is a problem with the user's equipment or a global problem of the building's pipe network. The judgment result is output in a standardized level form for subsequent operation and maintenance scheduling and handling. Specifically, this step compares the absolute value of the cumulative statistical deviation of the target user with a first preset threshold (e.g., 30%), and combines the deviation distribution of adjacent users on the same floor to make the following two types of judgments: First, if the absolute value of the target user's deviation is greater than or equal to the first preset threshold, while the absolute values ​​of the deviations of adjacent users on the same floor are all less than or equal to the second preset threshold (e.g., 10%), it means that the anomaly is limited to the current user and is unrelated to the building's pipe network. Therefore, it is judged as a local anomaly of a single household, and the first anomaly level (level one warning) is output, prompting operation and maintenance personnel to enter the household to check for pipe blockage, valve jamming, or thermostat failure. Secondly, if the absolute value of the deviation of the target user is greater than or equal to the first preset threshold, and the absolute value of the deviation of adjacent users on the same floor exceeding a preset proportion (e.g., 70%) is also greater than or equal to the third preset threshold (e.g., 20%), it indicates that the anomaly has affected the entire floor or the entire building, and is judged as a building-level global hydraulic imbalance. A second anomaly level (level two warning) is output, prompting dispatchers to perform overall pressure regulation or dredging of the building's pipe network. Figure 5 As shown, this solution can also use a unified color coding rule to visually identify the above-mentioned abnormal levels. Figure 5 This is a color-coded legend for anomaly levels. Green indicates normal operation, meaning the cumulative deviation of the target user and its neighbors does not exceed the first preset threshold. Yellow indicates the first anomaly level (Level 1 warning), corresponding to a local anomaly in a single household, prompting maintenance personnel to conduct on-site inspections. Red indicates the second anomaly level (Level 2 warning), corresponding to a building-wide hydraulic imbalance, prompting dispatch personnel to conduct a comprehensive inspection of the building's piping network. This color-coded legend allows maintenance personnel to quickly identify the type and urgency of the anomaly based on the warning color, thus matching the appropriate response procedure.

[0030] In this technical solution, after collecting the user's real-time heating flow rate, instead of directly using the absolute flow rate value as the basis for anomaly judgment, the relative deviation of the instantaneous flow rate per unit area is first calculated using the rated design flow rate and the heating building area, thereby eliminating numerical differences caused by different heating areas. Subsequently, a preset time window is activated, and multiple deviation values ​​within the window are accumulated and statistically analyzed to filter out transient interferences such as temporary water releases or water pressure fluctuations by users. Based on this, if the accumulated deviation exceeds the anomaly threshold, similar deviation data of adjacent users on the same floor are further obtained, and the anomaly of the target user is cross-validated using the deviation status of neighboring households. Since the neighboring households and the target user are on the same floor and on the same heating riser, their pipe network conditions are basically the same, and comparison can effectively distinguish between local faults in a single household and overall hydraulic imbalance in the building. In this way, the entire identification process no longer relies on the original absolute flow rate value or the instantaneous value at a single sampling moment, but undergoes multiple verifications through normalization, cumulative filtering, and neighborhood comparison, thereby ensuring the accuracy and reliability of anomaly judgment to a certain extent.

[0031] Based on the technical solutions described above, this specification also provides some specific implementation schemes, which are described below.

[0032] In an optional embodiment, the step of calculating the relative deviation of instantaneous flow rate per unit area based on the real-time heating flow rate, the rated design heating flow rate, and the heating building area may specifically include: Calculate the real-time flow rate per unit area by dividing the real-time heating flow rate by the heating building area; The design rated flow rate per unit area is calculated by dividing the rated design heating flow rate by the heating building area. The absolute deviation of the flow rate per unit area is calculated as the real-time flow rate per unit area minus the design rated flow rate per unit area. Divide the absolute deviation of the flow rate per unit area by the design rated flow rate per unit area and multiply by 100% to obtain the relative deviation of the instantaneous flow rate per unit area. The step of accumulating and statistically analyzing the relative deviations of instantaneous flow rate per unit area within a preset time window includes: setting a sliding time window of fixed duration; continuously collecting a preset number of relative deviations of instantaneous flow rate per unit area within the time window at a preset sampling period; calculating the arithmetic mean of the preset number of relative deviations of instantaneous flow rate per unit area as the accumulated deviation; wherein, the formula for calculating the accumulated deviation is: In the formula, The cumulative average normalized deviation over the time window. For the first in the window Instantaneous normalized flow deviation for each sampling period This represents the number of valid sampling points within the window. Take a preset positive integer; The step of accumulating and statistically analyzing the relative deviations of instantaneous flow rates per unit area within a time window with a preset duration includes: maintaining the time window using a mechanism of sliding and updating each sampling period; removing the oldest historical data within the window for each new sampled data, so that the window always maintains the latest preset number of continuous sampled data.

[0033] In this embodiment, the calculation process for the relative deviation of instantaneous flow rate per unit area involves a step-by-step normalization process, mapping user data from different heating areas to a unified dimensionless dimension. Specifically, the collected real-time heating flow rate is first divided by the user's heating building area to obtain the real-time flow rate per unit area. This value reflects the actual circulating flow rate per square meter of heating area under the current operating conditions. Simultaneously, the rated design heating flow rate is divided by the same heating building area to obtain the rated design flow rate per unit area, which serves as a benchmark reference for the user's normal operating conditions. Subtracting the two yields the absolute deviation of the flow rate per unit area. This deviation still carries dimensional information related to the design benchmark, so it needs to be further divided by the rated design flow rate per unit area and multiplied by 100% to finally obtain a dimensionless relative deviation expressed as a percentage. Through this step-by-step conversion process, the traffic status of users with different areas is unified under the same reference system for comparison. For example, a certain 100 The user's real-time traffic is 0.6. / h, with a design flow rate of 1.0 m³ / h, its normalized deviation is (0.6-1.0) / 1.0×100%=-40%; another 50 The user's real-time flow rate is 0.3 m³ / h, and the design flow rate is 0.5 m³ / h. Their normalization deviation is also (0.6-1.0) / 1.0×100%=-40%. After normalization, both show completely consistent deviation values, indicating that this indicator can completely eliminate the interference of heating area on flow rate determination. This provides a comparable data basis for setting a unified anomaly threshold for cross-user comparisons. Furthermore, since the heating building area is eliminated during the simplification process in the calculation formula, only the real-time flow rate and the design flow rate are needed for actual calculation. This simplifies the calculation process by providing a value.

[0034] Meanwhile, in this embodiment, the purpose of the time window cumulative statistical mechanism is to filter out the impact of minute-level random interference such as temporary water release by users and instantaneous fluctuations in water pressure on anomaly detection. Specifically, a fixed-length sliding time window can be set first. The window length can be determined based on the actual sampling period; for example, when the sampling period is 1 minute, the window contains deviation data from 5 consecutive sampling points. Then, using this sampling period as a step size, a preset number of relative deviations in instantaneous flow rate per unit area are continuously collected within the window, and the arithmetic mean of these deviations is calculated as the cumulative statistical result of the window. This averaging process smooths out abnormal fluctuations in individual sampling points, allowing the window output value to more accurately reflect the user's continuous operating status within that time period. Furthermore, window maintenance employs a sliding and updating mechanism on a per-sampling-per-period basis. This means that for each new deviation data point, the oldest historical data point within the window is removed, ensuring the window always maintains the latest N consecutive sampling data. For example, if the deviations of five sampling points within a minute window are -5%, -38%, -6%, -4%, and -7%, their arithmetic mean is -12%, far below the abnormal threshold. Even if the instantaneous value of -38% exceeds the threshold, no alarm will be triggered. This effectively avoids false alarms caused by single-point instantaneous interference, making the final judgment result more robust and reliable.

[0035] In an optional embodiment, the step of activating a preset time window and cumulatively calculating the relative deviations of instantaneous flow rates per unit area within the time window may further include: Calculate the stability index of the relative deviation sequence of instantaneous flow rate per unit area within the time window. and trend change rate : In the formula, Characterizing the instantaneous fluctuation intensity of the deviation sequence, The rate of absolute value deterioration of the biased sequence; When the stability index The trend change rate is less than the preset stability threshold and the trend change rate is less than the preset stability threshold. When the current anomaly exceeds the preset deterioration threshold, it is determined to be a gradual traffic anomaly fault, and the length of the time window is dynamically reduced to a preset proportion of the original length to accelerate the anomaly response. When the stability index When the current anomaly is greater than or equal to the preset stability threshold, it is determined to be a transient disturbance or a sudden fault, and the original length of the time window is maintained.

[0036] In this embodiment, the aforementioned fixed-duration sliding time window applies a uniform filtering scale to all operating conditions. However, in actual heating scenarios, the time series variation patterns of deviation data for different types of anomalies exhibit fundamental differences. For example, for faults involving gradually worsening blockages, the flow deviation often shows a continuous and slow deterioration trend, with the absolute value of the deviation monotonically increasing over multiple sampling periods. The instantaneous jumps between adjacent sampling points are relatively small, resulting in a generally stable deviation sequence. Conversely, instantaneous pressure disturbances or temporary water releases by users manifest as sharp jumps in the deviation at individual moments, but there is a lack of sustained directional consistency between preceding and following sampling points, leading to drastic fluctuations in the deviation sequence without a stable trend. Based on this physical difference, this solution calculates two additional auxiliary indicators within the time window to distinguish between the two types of operating conditions: stability indicators. Calculate the mean absolute value of the deviation change between adjacent sampling points to reflect the instantaneous fluctuation intensity of the deviation sequence. A smaller value indicates a smoother change in the deviation values ​​at each sampling point and a more stable sequence trend. A larger value indicates that the bias jumps frequently between different sampling times and lacks continuity. Trend change rate This only accumulates the changes in the absolute value of the deviation among adjacent sampling points, reflecting the rate of continuous deterioration of the absolute value of the deviation within the window. A larger value indicates that the deviation is in a continuous deepening process over time, rather than a random deviation that recovers immediately. These two indicators can characterize the dynamic features of the deviation sequence from different perspectives. The focus is on determining whether the sequence is stationary. The focus is on determining whether the sequence exhibits a unidirectional deterioration trend. Using both methods in combination can more clearly distinguish the formation mechanism of the deviation. The stability threshold, the deterioration threshold, and the preset ratio are all pre-set empirical parameters that can be calibrated using the statistical distribution characteristics of historical operating data of the heating system. In an exemplary, non-limiting, embodiment, the stability threshold can be set to any value within the range of 5% to 10% (e.g., 5%), the deterioration threshold can be set to any value within the range of 2% / min to 5% / min (e.g., 3% / min), and the preset ratio can be set to any value within the range of 60% to 80% (e.g., 60%, i.e., the window length dynamically shrinks from 5 minutes to 3 minutes). The specific values ​​of the above thresholds and ratios can be adaptively adjusted according to the network characteristics, instrument accuracy, and historical anomaly samples of different heating systems, and can be determined by those skilled in the art through a limited number of experiments or statistical regression analysis.

[0037] Based on the numerical combination of the two indicators mentioned above, this scheme also dynamically adjusts the time window length to achieve differentiated responses to different anomaly patterns. Specifically, when the stability index... The rate of change is less than the preset stability threshold, and the trend change rate is... When the deviation exceeds the preset deterioration threshold, it indicates that the deviation sequence within the current time window changes smoothly without significant jumps, and the absolute value of the deviation is continuously increasing. This combination of characteristics corresponds to a typical gradual flow anomaly fault, such as a slow decrease in flow caused by gradual scaling on the inner wall of the pipeline or a valve gradually closing due to mechanical wear. For this type of anomaly, the deviation trend has a clear trend and is predictable, and can be identified without accumulating too many data points. Therefore, this solution dynamically reduces the length of the time window to a preset proportion of the original length, such as shortening it to 3 minutes, so that the window contains only a few sampling points to trigger subsequent judgments. This can shorten the waiting time for anomaly confirmation and improve the early warning response speed while ensuring reliability. Conversely, when the stability index... When the deviation is greater than or equal to the preset stability threshold, it indicates that there are significant jumps in the deviation sequence at each sampling time, and the sequence trend is unstable. At this time, regardless of whether the absolute value of the deviation exceeds the abnormal threshold, it cannot be confirmed that the deviation has a continuous development trend. It may be caused by instantaneous interference or sudden impact. Therefore, the original length of the time window is kept unchanged, and a sufficient number of sampling points are accumulated for mean filtering to fully smooth the fluctuations before entering the subsequent judgment stage. Through this adaptive window adjustment mechanism based on the dynamic characteristics of the deviation sequence, this scheme can enable the fixed-length time window to be extended in real time according to the working conditions. It can respond quickly to slowly deteriorating gradual faults, and maintain the original filtering strength for instantaneous fluctuations without introducing additional risks. This avoids the problem of response lag or insufficient filtering caused by "one-size-fits-all" application of a uniform window length when facing different types of anomalies. Figure 4As shown in the figure, this is a schematic diagram of the propagation process of a gradual flow anomaly fault, demonstrating the complete evolution of the fault from its initial stage to its eventual rapid expansion. Stage T1 represents the initial stage of the fault, where the deviation is still within the normal range, and the response speed before and after window adjustment is not significantly different. Stage T2 represents the fault propagation stage, where the deviation begins to deviate from the normal range and gradually approaches the anomaly threshold; the adjusted, shortened window can capture this deviation trend earlier. Stage T3 represents the rapid expansion stage, where the deviation is close to the anomaly threshold, and the adjusted window reaches the warning trigger point earlier than the original window. Through this dynamic adjustment mechanism, the time window is no longer a fixed length but adaptively shrinks or remains constant according to the inherent changing characteristics of the deviation sequence. This allows the system to maintain sufficient filtering capability to suppress random interference during steady-state operation and automatically improves response sensitivity when the fault worsens, balancing anti-interference capability and timeliness.

[0038] In an optional embodiment, in the step of obtaining the relative deviation of instantaneous flow per unit area of ​​adjacent users on the same floor, the adjacent users are a preset number of users who are pre-selected, are on the same floor as the target user, are on the same heating riser, and whose heating area deviation does not exceed a preset ratio. The step of obtaining the relative deviation of instantaneous flow per unit area of ​​adjacent users on the same floor when the absolute value of the cumulative statistical deviation exceeds the preset abnormal threshold is as follows: when the absolute value of the cumulative statistical deviation of a preset number of consecutive sampling periods exceeds the first preset threshold, cross-validation of neighboring households is triggered to obtain the relative deviation of instantaneous flow per unit area of ​​adjacent users on the same floor.

[0039] In this embodiment, to ensure the comparability of neighboring household data and the effectiveness of cross-validation during actual deployment, static screening of adjacent reference objects for each household is performed in advance. Specifically, for a target user, several households on the same floor, sharing the same heating riser, and with heating area deviations not exceeding a preset ratio are pre-selected as fixed neighboring household samples. The sample size is a preset number of households (e.g., 3 to 5 households). The emphasis on "same floor" and "same heating riser" in this solution is because the heating network exhibits significant hydraulic stratification in the vertical direction. The water pressure and flow distribution between different floors and different risers often differ considerably. Including users from different floors or risers as references could introduce non-faulty operational interference, leading to misjudgments. The requirement that "heating area deviations not exceed a preset ratio" ensures that the baseline flow rates of each household are on the same order of magnitude before normalization, thereby improving the fairness and accuracy of subsequent deviation comparisons. (Appendix) Figure 2This diagram illustrates the user distribution across different floors and risers within a building. The target user is located on the 3rd floor and belongs to riser number 2. Users on the same floor (3rd floor) and sharing riser number 2 are labeled as neighbor 1 and neighbor 2, respectively. Users on the same floor but belonging to different risers (e.g., riser 4) are not included in the target user's reference sample. This method minimizes interference from differences in the network topology. After statically screening neighbor samples, specific conditions for triggering neighbor cross-validation need to be set to avoid frequent unnecessary neighbor comparisons due to occasional data jumps. This scheme can use the condition that the cumulative deviation after a continuous preset sampling period exceeds a first preset threshold as the trigger condition. Specifically, the sampling period can be set to 1 minute, and the cumulative statistics refer to the window average normalized deviation calculated using a 5-minute sliding time window (i.e., as mentioned above). When the average deviation of the window for a consecutive preset number of sampling periods (e.g., 3 consecutive periods) is greater than or equal to the first preset threshold (e.g., 30%), it is considered that the target user is indeed in a continuous abnormal state, rather than caused by a momentary disturbance. At this time, the neighboring household cross-validation process is carried out to obtain the average deviation data of each pre-selected neighboring household within the same time window.

[0040] After confirming the persistent anomaly of the target user based on the above conditions, the anomaly type can be accurately distinguished and its level output can be determined based on the deviation data of neighboring households. The specific judgment rule is as follows: if the absolute value of the target user's cumulative average normalized deviation within a 5-minute sliding time window is greater than or equal to the first preset threshold (i.e., the anomaly judgment baseline, e.g., 30%), and the absolute value of the cumulative average normalized deviation of all pre-selected neighboring users on the same floor within the same time window is less than or equal to the second preset threshold (e.g., 10%), then it is judged as a single-household local anomaly. Under this condition, the heating status of the neighboring households is within the normal range, indicating that the target user's deviation is not caused by the overall hydraulic imbalance of the building's pipe network, but rather by a problem with its own heating equipment or pipes. The source of the fault can be further located as a local equipment anomaly such as blockage of the inlet pipe, stuck valve at the household end, or malfunctioning thermostat. At this time, the first anomaly level (i.e., level one warning) is output, prompting maintenance personnel to conduct on-site inspection of this specific user. Conversely, if the absolute value of the cumulative average normalized deviation of the target user is still greater than or equal to the first preset threshold, and the absolute value of the cumulative average normalized deviation of adjacent users on the same floor exceeding a preset proportion (e.g., 70%) within the same time window is also greater than or equal to the third preset threshold (e.g., 20%), then it is determined to be a building-level global hydraulic imbalance. This condition indicates that the abnormal flow is not unique to a single household, but rather a general flow deviation across the entire floor or building. The root cause lies in systemic hydraulic problems such as uneven pressure distribution in the building-level pipe network, throttling of the main pipe, or pipe blockage. At this time, a second abnormality level (i.e., a level two warning) is output, prompting dispatchers to perform overall pressure regulation or pipe unblocking of the building's pipe network. It should be noted that the second and third preset thresholds differ numerically. The second preset threshold (e.g., 10%) represents the upper limit of tolerance for neighboring households being completely in normal operating conditions, while the third preset threshold (e.g., 20%) represents the minimum threshold for determining when neighboring households have generally entered an abnormal state. The numerical gap between the two ensures that there is no ambiguity in the determination between local anomalies and global imbalances, thereby guaranteeing the clarity and reliability of anomaly classification. Through the above two-layer determination logic, this solution separates single-household equipment failures from building system failures, providing a clear basis for the precise scheduling of subsequent operation and maintenance resources, and avoiding blind repairs or invalid work orders caused by the inability to distinguish fault types in traditional solutions.

[0041] In an optional embodiment, the step of identifying the abnormal state and outputting the corresponding abnormal level based on the comparison result of the relative deviation of the instantaneous flow per unit area of ​​the adjacent users with the preset abnormal threshold may specifically include: If the absolute value of the cumulative statistical deviation of the target user is greater than or equal to the first preset threshold, and the absolute value of the cumulative statistical deviation of its adjacent users in the same layer is less than or equal to the second preset threshold, then it is determined to be a single-household local anomaly, and the first anomaly level is output. If the absolute value of the cumulative deviation of the target user is greater than or equal to the first preset threshold, and the absolute value of the cumulative deviation of the adjacent users on the same floor exceeding a preset proportion is greater than or equal to the third preset threshold, then it is determined to be a building-level global hydraulic imbalance, and the second abnormal level is output.

[0042] In the specific implementation process, this solution pre-sets three thresholds for anomaly detection: a first preset threshold, a second preset threshold, and a third preset threshold, and stipulates a preset proportion (e.g., 70%) for triggering global detection in adjacent user groups. The first preset threshold is used to evaluate whether the target user's own flow deviation reaches an abnormal level. The second and third preset thresholds are used to measure the upper tolerance limit and lower trigger limit of deviation for neighboring households in scenarios of local anomaly and global imbalance, respectively. As a typical, but not limiting, embodiment, the first preset threshold can be 30%, the second preset threshold can be 10%, the third preset threshold can be 20%, and the preset proportion can be 70%. When the absolute value of the normalized deviation of the target user, obtained after averaging over a 5-minute sliding time window, is greater than or equal to 30%, it indicates that the user's heating flow has continuously deviated from the design value, meeting the anomaly triggering condition. At this point, the window average deviation data of 3 to 5 neighboring households on the same floor, with the same riser, and similar area, which were pre-selected, are immediately retrieved. If the absolute value of the deviation for these neighboring households is less than or equal to 10%, it indicates that the heating status of these neighboring households is basically normal, and their flow deviation is within the daily fluctuation range without any obvious continuous shift. However, the target user shows a significant decrease or increase in flow, suggesting that the fault source is limited to the target user's own indoor equipment, such as partial blockage of the inlet branch pipe, stuck valve at the household end, malfunctioning thermostatic valve, or clogged filter. This type of anomaly will not have a substantial impact on the heating effect of other users on the same floor. Therefore, this solution classifies it as a localized anomaly in a single household and outputs the first anomaly level (e.g., Level 1 warning), with the corresponding maintenance instruction being to conduct an on-site inspection and repair of the household's equipment. Conversely, if the absolute value of the target user's average window deviation is also greater than or equal to 30%, but statistics show that more than a preset proportion (e.g., 70%) of neighboring households on the same floor also have an absolute value of 20% or more in their average window deviation, it indicates that the deviation is not unique to the target user, but rather that multiple users on the entire floor or even the entire building are showing the same trend of flow deviation. Because users on the same riser or floor share similar inlet pressure and main water supply conditions, when multiple neighboring households experience significant deviations simultaneously, the root cause is usually not within a single household, but rather a problem with the building-wide water supply system. This could be a systemic hydraulic imbalance such as a clogged building inlet filter, incorrect main regulating valve settings, localized throttling in the riser, or an imbalance in pump outlet pressure distribution. In this case, it is classified as a building-wide hydraulic imbalance, and a second-level anomaly (e.g., a level-two warning) is issued, prompting maintenance personnel to conduct a comprehensive inspection, pressure adjustment, or unblocking of the building's pipe network, rather than performing individual household repairs. Through this hierarchical judgment logic, this solution can accurately distinguish between localized individual faults and global systemic faults, avoiding the embarrassing situation of dispatching work orders to households due to misjudgment but failing to resolve the problem, thereby significantly improving the efficiency of heating operation and maintenance. Figure 3As shown in the figure, this is a comparison table of the instantaneous flow deviation distribution per unit area for each user under a building-level global hydraulic imbalance scenario, displaying the user deviation distribution for different risers and different units in a certain building. The horizontal axis represents the riser numbers (risers 1 to 4), and the vertical axis represents the unit location (east unit, middle unit, west unit). The values ​​in the table are the cumulative average deviation percentages for each user after normalization and time window filtering. It can be clearly seen from the figure that the deviations of the three units located on riser 2 (east unit -45.0%, middle unit -44.8%, west unit -43.9%) are highly consistent, all approximately -44%, with deviations not exceeding 1.5 percentage points from each other, and all far exceeding the third preset threshold of 20%. Meanwhile, the deviations for all users under other risers (risers 1, 3, and 4) in the same building are around -4% to -5%, within the normal operating range. This data pattern indicates that the deviation is not isolated to a single user, but rather a synchronous flow shortage affecting all users served by the entire riser, representing a typical riser-level global hydraulic imbalance. Figure 3 The specific data shown demonstrates that this solution can quickly identify such global faults in actual operation and maintenance, avoiding the misjudgment of systemic problems as local anomalies due to focusing only on individual user data.

[0043] In optional embodiments, the technical solution may further include the following steps: Construct an anomaly severity index The calculation formula is as follows: In the formula, The deviation is calculated cumulatively for the target users. The first preset threshold, The first nonlinear weighting coefficient, The number of adjacent users, For the first The deviation after cumulative statistics of adjacent users The normalized benchmark threshold for the deviation level between neighboring households can be dynamically determined based on the anomaly judgment scenario, i.e., in the single-household local anomaly judgment scenario (i.e., the target user's deviation is triggered while the neighboring household's deviation is normal, corresponding to the judgment situation of the first anomaly level mentioned above). Take the second preset threshold. In the scenario of determining global hydraulic imbalance at the building level (i.e., there is a general deviation among neighboring households, corresponding to the determination of the second abnormal level mentioned above). Take the third preset threshold. The second nonlinear weighting coefficient, The standard deviation of the cumulative deviations of all adjacent users. This is the deviation dispersion adjustment coefficient; According to the severity index of the anomaly The preset value range in which the first or second abnormality level is located is used to refine the first or second abnormality level into multiple sub-levels, and corresponding sub-level warning signals are output.

[0044] In this embodiment, to further improve the granularity of anomaly severity differentiation and enable maintenance personnel to more intuitively judge the urgency of the fault, this solution constructs an anomaly severity index. This index nonlinearly fuses the target user's own deviation magnitude, the deviation level of neighboring households, and the dispersion of deviation among neighboring households to provide a continuous, quantifiable measure of severity. Specifically, It consists of the product of three factors, where the first factor is... Characterizes the severity of the target user's own bias. As mentioned above, the first preset threshold (i.e., the baseline for determining the existence of anomalies) is used when the average window deviation of the target user... When the baseline is exceeded, this factor is greater than 1, and the greater the exceedance, the more non-linear the increase in value, amplifying the severity. The second factor... Characterizing the average deviation level of neighboring households, The value selection rule is as described in the parameter explanation section above. It dynamically selects the corresponding threshold based on the anomaly scenario currently affecting the target user. For example, a second preset threshold (e.g., 10%) is used in local anomaly scenarios, and a third preset threshold (e.g., 20%) is used in global imbalance scenarios. This ensures that the factor is always normalized based on a benchmark matching the current anomaly type. When the average deviation of neighboring households exceeds the corresponding threshold, the factor is greater than 1, reflecting that the fault has spread from a single household to surrounding users, increasing the global risk. Conversely, if the deviation of neighboring households is generally low, the factor is less than 1, reducing the severity. The third factor... This introduces the standard deviation of the neighboring household deviation. As a dispersion adjustment term, when the deviation data of neighboring households differ significantly. As the error rate increases, this factor also increases, indicating the presence of uneven distribution of anomalies within the system, increasing overall complexity and difficulty of handling. Conversely, if the deviations of neighboring households are highly consistent, the factor approaches 1 and does not amplify the severity. The result is obtained by multiplying the three factors. It is a dimensionless continuous value that can comprehensively reflect information from three dimensions: the depth of a single household's fault, the range of influence in the neighborhood, and the system's non-uniformity, providing a quantitative basis for subsequent refined classification.

[0045] To facilitate understanding of the interaction logic of the above parameters, a specific example is provided below: Suppose that the average window deviation of a target user is... The first preset threshold is 45%. It is 30%, take The first factor is then calculated as Simultaneously, the absolute deviation values ​​of three neighboring households were collected, which were 8%, 42%, and 46%, respectively. Since multiple households among the neighboring households showed significantly excessive deviations, this scenario falls under the category of building-level global hydraulic imbalance assessment. The third preset threshold of 20% should be taken, then the deviation ratio of the neighboring households' average value is 10%. ,Pick The second factor is approximately Standard deviation of the deviation among three neighboring households Approximately 17.6%, taken The third factor is Multiplying the three together gives In contrast, if the second preset threshold of 10% is incorrectly used as the threshold for local anomalies... This would result in a deviation multiplier of 3.2 and a second factor of 2.45. The score of 4.17 significantly exaggerates the severity of the anomaly. This solution dynamically selects criteria that match the judgment scenario. The selection of a specific value can avoid misjudging the severity due to an inappropriate threshold selection. Through the aforementioned dynamic threshold matching mechanism, It can objectively reflect the true severity of anomalies in the current judgment scenario, providing an accurate basis for subsequent refined classification.

[0046] In obtaining After obtaining the specific numerical values, the original first or second anomaly level can be further refined into multiple sub-levels based on several pre-defined numerical ranges. Specifically, the original two-level classification—the first anomaly level corresponding to localized anomalies in a single household, and the second anomaly level corresponding to overall hydraulic imbalance at the building level—is essentially still a qualitative classification and cannot distinguish the severity and urgency within the same type of anomaly. However, through… The continuous numerical characteristics allow for the setting of multiple sub-level intervals, such as mild, moderate, and severe, with each interval corresponding to a specific [specific level]. Numerical range. After the anomaly type determination (local or global) is completed, the value under the current operating condition is calculated immediately. The value is calculated and its corresponding preset range is determined, thus outputting the corresponding sub-level warning signal. Different sub-levels can then correspond to different maintenance response strategies. Specifically, for example, a mild sub-level only needs to be included in the daily inspection plan, a moderate sub-level needs to be scheduled for maintenance on the same day, and a severe sub-level triggers immediate work order processing. Continuing with the previous example, after correctly selecting... The result was obtained after calculation. If it falls into the system's preset In the moderate range, the anomaly level corresponding to the target user is marked as a moderate sub-level. Maintenance personnel can arrange repairs on the same day without immediate work order dispatch, thus achieving a reasonable allocation of maintenance resources. Through this quantitative grading mechanism, this solution transforms the originally discrete anomaly judgment results into continuous risk metrics. This allows maintenance personnel to understand the severity of the fault immediately upon receiving an alert, making preliminary priority judgments without additional analysis of raw data, thereby improving the efficiency and accuracy of anomaly response.

[0047] In optional embodiments, the technical solution may further include the following steps: Calculate the spatial dispersion index of the deviation between the target user and all its neighboring users. and homogeneity coefficient : In the formula, The number of adjacent users, For the first The deviation after cumulative statistics of adjacent users The mean of the absolute values ​​of the deviations between adjacent users. Use extremely small positive numbers to prevent division by zero errors; Wherein, the global imbalance interval, the intermediate interval, the first homogeneity threshold, the upper limit of dispersion, and the lower limit of dispersion are all pre-set empirical judgment boundaries, which can be calibrated using historical statistical data of neighboring household deviations under normal operating conditions of the heating system. In an exemplary and non-limiting embodiment, the global imbalance interval can be set as follows: The intermediate interval can be set as follows: The first homogeneity threshold can be set to: (Right now When the deviation of the target user is significantly higher than the average of neighboring households (when the deviation is determined to be significant), the upper limit of the dispersion can be set to The lower limit of the dispersion can be set to: (Right now The deviations of neighboring households were highly consistent. (There is significant differentiation in the deviation among neighboring households). The specific values ​​of the above thresholds and interval boundaries can be adaptively adjusted according to the pipeline topology of different heating systems, the number of neighboring household samples, and the characteristics of historical abnormal distributions. Those skilled in the art can determine these values ​​through a limited number of experiments or statistical regression analysis.

[0048] In this embodiment, to further improve the precision of anomaly type determination, two auxiliary statistics are introduced on top of the basic neighbor cross-validation logic: the deviation spatial dispersion index. and homogeneity coefficient Among them, the spatial dispersion index of the deviation The standard deviation is the absolute value of the deviations of all neighboring users. It measures the dispersion of deviations among neighboring households. A smaller value indicates that the deviations of neighboring households are more similar and the spatial distribution of deviations is more uniform. A larger value indicates that the deviation levels among neighboring households are uneven, with some households showing significantly higher or lower deviations, indicating differentiation within the group. Homogeneity coefficient This is the ratio of the absolute value of the target user's deviation to the mean of the absolute values ​​of the deviations of all neighboring households. It measures the degree to which the target user's deviation is more pronounced relative to the neighboring household group. A larger value indicates that the target user's deviation is significantly higher than the average level of the neighboring households, while a smaller value indicates that the target user's deviation is not significantly higher than the general level of the neighboring household group. These two indicators can characterize the relative relationship between the target user and the neighboring household group from different dimensions. The focus is on determining whether the target user is an "outlier" within the group, while The focus is on determining whether the neighboring households themselves exhibit consistency. Only... And lack The constraints may lead to situations where the target user's deviation is much greater than the neighboring household's mean, even though there is already a large dispersion among the neighboring households. In such cases, simply classifying the target user as an isolated anomaly is insufficient; conversely, only... And lack The constraints of the previous method also fail to distinguish the essential difference between consistently low and consistently normal levels in neighboring households. Therefore, the technical solution of this embodiment uses the two indicators together to form a two-dimensional judgment space, which can provide a more comprehensive basis for subsequent three-branch anomaly type classification.

[0049] Based on the numerical combination of the above two indicators, the technical solution of this embodiment also employs a three-branch logic to refine the classification of anomaly types. Specifically, in the first branch, when the homogeneity coefficient... Falling into a preset global imbalance range (for example, as described above) Meanwhile, the spatial dispersion index Less than the preset upper limit of dispersion (e.g., as described above) When the homogeneity coefficient is significantly lower than the average deviation of neighboring households, it indicates that the absolute value of the deviation of the target user is significantly smaller than that of the neighboring households. This means that the deviation of the neighboring households is generally higher than that of the target user, and the deviations among the neighboring households are highly consistent with each other, showing almost no dispersion. This condition corresponds to the situation where most users in the entire building or floor have a unidirectional flow deviation, while the target user has a relatively mild deviation. The root cause is that the building-level hydraulic imbalance has a general impact on all users. Therefore, the fault type is determined to be a building-level global hydraulic imbalance. Furthermore, because the deviation has a wide impact and involves many users under this condition, the original second-level anomaly is upgraded to a first-level warning priority to emphasize the urgency of its handling. Specifically, in the second branch, when the homogeneity coefficient... Greater than or equal to the first homogeneity threshold (e.g., as described above) Meanwhile, the spatial dispersion index Less than the preset upper limit of dispersion (e.g., as described above) When the homogeneity coefficient is [value missing], it indicates that the absolute value of the target user's deviation is significantly higher than the average level of the neighboring households, while the deviations among the neighboring households are highly consistent and at a low level. This suggests that the target user is the only individual among the surrounding users exhibiting a significant deviation, and the neighboring households as a whole are normal. Therefore, the fault type is determined to be a single-household local anomaly, maintaining the first anomaly level output without further increasing the warning priority. Specifically, in the third branch, when the homogeneity coefficient [value missing]... Falling into the preset middle range (for example, as described above) Meanwhile, the spatial dispersion index Greater than the preset lower limit of dispersion (e.g., as described above) When the deviation of the target user is similar to the average level of the neighboring households, but the deviation levels of each neighboring household differ significantly and the dispersion within the group is high, it indicates that it is not an isolated anomaly of a single household, nor is it a global imbalance with all users exhibiting a highly uniform direction. Instead, it suggests that multiple users have independently experienced local anomalies of varying degrees, and there is no unified hydraulic root cause among these anomalies. Therefore, the fault type is determined to be the superposition of multiple isolated local anomalies, with each user maintaining its own first anomaly level output. Through the above three-branch logic, the technical solution of this embodiment expands the original two-category judgment that only distinguishes between local anomalies and global imbalances to three categories. It can add the ability to identify the intermediate state of "group dispersion but not global uniform direction," thereby avoiding the situation where the dispersion of the neighboring household group is too high, leading to a misjudgment of global imbalance. This makes the anomaly classification more consistent with the complex and variable fault distribution patterns in actual heating systems.

[0050] In optional embodiments, the technical solution may further include the step of: collecting the supply and return water temperature difference of the user's target heating facility or building as a multi-dimensional feature to assist in verifying the abnormal state.

[0051] In the embodiments of this specification, the supply and return water temperature difference refers to the temperature difference between the supply water temperature and the return water temperature in the heating system. According to the basic principles of thermodynamics, there is a clear physical relationship between the heat released on the user side, the flow rate, and the supply and return water temperature difference. That is, when the user's heat load remains stable, the flow rate is inversely proportional to the supply and return water temperature difference. In other words, when an abnormality such as pipe blockage or valve jamming causes a decrease in flow rate, the hot water stays on the user side for a longer time and dissipates heat more fully, and the supply and return water temperature difference will increase accordingly; conversely, when an equipment failure causes a sudden increase in flow rate, the supply and return water temperature difference will decrease. Therefore, the introduction of this multi-dimensional feature of supply and return water temperature difference can complement the flow rate deviation for verification. That is, if the flow rate deviation is negative (insufficient flow) and the supply and return water temperature difference increases simultaneously, the credibility of blockage-related faults is significantly improved. If the flow rate deviation is negative but the supply and return water temperature difference is normal or the trend of change does not match the flow rate, it may be caused by non-equipment failure factors such as flow meter zero-point drift or pipeline pressure fluctuations. Therefore, by collecting the temperature difference data and performing time-series correlation analysis with the flow deviation in this embodiment, misjudgments caused by instrument errors and external disturbances can be effectively eliminated, further improving the accuracy and reliability of anomaly diagnosis.

[0052] Based on the technical solution of the previous embodiment, the step of identifying the abnormal state and outputting the corresponding abnormal level according to the comparison result of the relative deviation of the instantaneous flow per unit area of ​​the adjacent users and the preset abnormal threshold may further include: Obtain the relative deviation sequence of instantaneous flow per unit area for the target user within the time window. and the corresponding supply and return water temperature difference sequence ; Calculate the time coupling coefficient between the deviation sequence and the supply and return water temperature difference sequence. : In the formula, The arithmetic mean of the supply and return water temperature difference within the time window; Based on the absolute value of the timing coupling coefficient The first preset threshold, the second preset threshold, and the third preset threshold are dynamically adjusted to obtain the adjusted dynamic threshold. The first preset threshold, the second preset threshold, and the third preset threshold are replaced with the corrected dynamic threshold, and the anomaly determination and level output are re-executed.

[0053] In this embodiment, to further improve the reliability of anomaly detection and reduce misjudgments caused by insufficient data from a single traffic dimension, a temporal coupling coefficient is introduced based on multi-dimensional feature supplementation. This method is used to quantitatively assess the synergistic relationship between the relative deviation of instantaneous flow rate per unit area and the supply and return water temperature difference within a time window. In this embodiment, the supply and return water temperature difference is considered a crucial parameter reflecting the thermal balance of a heating system. Specifically, when a fault such as blockage or valve stagnation occurs on the user side, leading to a decrease in flow rate, the supply and return water temperature difference typically increases accordingly with the decrease in flow rate, while the heating output remains essentially unchanged. The two exhibit an inverse coupling characteristic. Conversely, if the flow rate deviation originates solely from non-fault factors such as water temperature fluctuations or instantaneous pressure disturbances, there is often no stable synchronous change pattern between the temperature difference and flow rate, or the coupling direction is opposite to the expected direction. Based on this physical mechanism, this solution simultaneously collects two data sets within the time window: the normalized deviation sequence. and supply and return water temperature difference sequence And calculate the Pearson correlation coefficient between the two. The closer the absolute value of this coefficient is to 1, the more synchronized the temporal trends of the deviation and temperature difference are; that is, a decrease in flow rate is accompanied by a synchronous increase in temperature difference, or vice versa. The closer the value is to 0, the less obvious the linear correlation between the two data sets, and the more likely the flow deviation originates from non-faulty random disturbances. For example, in five sets of data collected from a target user within a 5-minute window, the normalized deviations are -32%, -35%, -33%, -36%, and -34%, respectively. The corresponding supply and return water temperature differences are +0.8℃, +0.9℃, +0.7℃, +1.0℃, and +0.8℃. The inverse trends of these two deviations are highly consistent. Therefore, the calculated... The value is close to -1. A value close to 1 indicates a strong coupling relationship between the decrease in flow rate and the increase in temperature difference, making the fault characteristics very clear. However, if the deviation also exceeds the threshold within adjacent time windows, but the temperature difference data shows no obvious regular change, for example, fluctuating around -30% while the temperature difference only fluctuates randomly with small jumps of 0.1℃, -0.1℃, 0.2℃, -0.2℃, and 0.1℃, then... When the value is close to 0, the decrease in flow rate is more likely due to pressure fluctuations or instantaneous errors in the flow meter, and should not be directly identified as a continuous heating failure.

[0054] Obtaining the timing coupling coefficient Subsequently, this scheme also utilizes its absolute value. The aforementioned first, second, and third preset thresholds are dynamically adjusted. This adjustment step is introduced because the reliability and fault indication of traffic data differ fundamentally under different operating conditions; that is, when… When the values ​​are high, it indicates that the flow deviation and changes in thermodynamic parameters corroborate each other, with multiple features pointing to the same anomaly source. In this case, the confidence level of the judgment is high, and the judgment thresholds can be appropriately lowered, allowing the system to trigger warnings even with slight deviations, thus improving sensitivity and the lead time for warnings. However, when... When the threshold is low, it indicates a lack of thermal data to support the flow deviation. The flow data alone is insufficient to confirm a genuine fault. In this case, the threshold should be appropriately raised to avoid false alarms caused by non-fault factors such as flowmeter drift or instantaneous pressure fluctuations. Specific correction methods can be designed as piecewise linear or nonlinear mappings. For example, when... When the condition is determined to be a strongly coupled working condition, the first preset threshold is adjusted downward from the original value of 30% to 25%, the second preset threshold is adjusted downward from 10% to 8%, and the third preset threshold is adjusted downward from 20% to 16%; when When the condition is determined to be a medium coupling condition, the thresholds are slightly lowered or remain at their original values; when When an anomaly is detected, a weakly coupled operating condition is identified, and the first preset threshold is adjusted upwards by 35%, with other thresholds adjusted proportionally. The resulting new thresholds are the dynamic judgment thresholds for the current operating condition. These revised thresholds then replace the original fixed thresholds, and the anomaly judgment and level output are re-executed. Through this dynamic threshold correction mechanism, this solution can quantitatively feed back information on the independent thermal dimension of the supply and return water temperature difference to the flow deviation judgment stage, achieving closed-loop verification between the flow and thermal sides. This effectively avoids misjudgment and missed judgment of a single flow index under multi-source interference, making the threshold setting no longer a fixed constant but a variable that adaptively adjusts according to the data coupling state, thereby further improving the robustness and adaptability of anomaly identification under different operating conditions.

[0055] It should be understood that in the methods described in one or more embodiments of this specification, the order of some steps may be adjusted according to actual needs, or some steps may be omitted.

[0056] Based on the foregoing technical solutions, this invention also provides a heating anomaly identification device based on instantaneous flow rate deviation per unit area, such as... Figure 6 As shown, the device, from a macroscopic perspective, may include the following modules: The data acquisition module 602 is used to collect the real-time heating flow rate, rated design heating flow rate and corresponding heating building area of ​​the target heating facility of the user in the heating system. The normalized deviation calculation module 604 is used to calculate the relative deviation of instantaneous flow per unit area based on the real-time heating flow, rated design heating flow, and heating building area. The sliding window cumulative filtering module 606 is used to start a time window of preset duration and accumulate and statistically analyze the relative deviations of instantaneous flow rates per unit area within the time window in order to filter instantaneous fluctuations. The neighbor data acquisition module 608 is used to acquire the relative deviation of instantaneous flow per unit area of ​​adjacent users on the same floor when the absolute value of the cumulative statistical deviation exceeds the preset abnormal threshold. The anomaly determination and classification output module 610 is used to identify anomalies and output corresponding anomaly levels based on the comparison result between the relative deviation of the instantaneous flow per unit area of ​​the adjacent users and the preset anomaly threshold. The formula for calculating the relative deviation of instantaneous flow rate per unit area is as follows: In the formula, The relative deviation of instantaneous flow rate per unit area. Real-time heat flow rate for the user's target heating facility. The rated design heating flow rate for the user's target heating facility.

[0057] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying heating anomalies based on instantaneous flow rate deviation per unit area, characterized in that, Includes the following steps: Collect the real-time heating flow rate, rated design heating flow rate, and corresponding heating building area of ​​the target heating facilities of users in the heating system. Based on the real-time heating flow rate, the rated design heating flow rate, and the heating building area, calculate the relative deviation of the instantaneous flow rate per unit area; A preset time window is started, and the relative deviations of instantaneous flow rates per unit area within the time window are accumulated and statistically analyzed to filter out instantaneous fluctuations; When the absolute value of the cumulative deviation exceeds the preset abnormal threshold, the relative deviation of the instantaneous flow per unit area of ​​adjacent users on the same floor is obtained. Based on the comparison between the relative deviation of the instantaneous flow per unit area of ​​the adjacent users and the preset abnormal threshold, the abnormal state is identified and the corresponding abnormal level is output. The formula for calculating the relative deviation of instantaneous flow rate per unit area is as follows: In the formula, The relative deviation of instantaneous flow rate per unit area. Real-time heat flow rate for the user's target heating facility. The rated design heating flow rate for the user's target heating facility.

2. The heating anomaly identification method based on instantaneous flow rate deviation per unit area as described in claim 1, characterized in that, The step of calculating the relative deviation of instantaneous flow rate per unit area based on the real-time heating flow rate, the rated design heating flow rate, and the heating building area specifically includes: Calculate the real-time flow rate per unit area by dividing the real-time heating flow rate by the heating building area; The design rated flow rate per unit area is calculated by dividing the rated design heating flow rate by the heating building area. The absolute deviation of the flow rate per unit area is calculated as the real-time flow rate per unit area minus the design rated flow rate per unit area. Divide the absolute deviation of the flow rate per unit area by the design rated flow rate per unit area and multiply by 100% to obtain the relative deviation of the instantaneous flow rate per unit area. The step of accumulating and statistically analyzing the relative deviations of instantaneous flow rate per unit area within a preset time window includes: setting a sliding time window of fixed duration; continuously collecting a preset number of relative deviations of instantaneous flow rate per unit area within the time window at a preset sampling period; calculating the arithmetic mean of the preset number of relative deviations of instantaneous flow rate per unit area as the accumulated deviation; wherein, the formula for calculating the accumulated deviation is: In the formula, The cumulative average normalized deviation over the time window. For the first in the window Instantaneous normalized flow deviation for each sampling period This represents the number of valid sampling points within the window. Take a preset positive integer; The step of accumulating and statistically analyzing the relative deviations of instantaneous flow rates per unit area within a time window with a preset duration includes: maintaining the time window using a mechanism of sliding and updating each sampling period; removing the oldest historical data within the window for each new sampled data, so that the window always maintains the latest preset number of continuous sampled data.

3. The heating anomaly identification method based on instantaneous flow rate deviation per unit area as described in claim 2, characterized in that, The step of accumulating and statistically analyzing the relative deviations of instantaneous flow rates per unit area within a preset time window further includes: Calculate the stability index of the relative deviation sequence of instantaneous flow rate per unit area within the time window. and trend change rate : In the formula, Characterizing the instantaneous fluctuation intensity of the deviation sequence, The rate of absolute value deterioration of the biased sequence; When the stability index The trend change rate is less than the preset stability threshold and the trend change rate is less than the preset stability threshold. When the current anomaly exceeds the preset deterioration threshold, it is determined to be a gradual traffic anomaly fault, and the length of the time window is dynamically reduced to a preset proportion of the original length to accelerate the anomaly response. When the stability index When the current anomaly is greater than or equal to the preset stability threshold, it is determined to be a transient disturbance or a sudden fault, and the original length of the time window is maintained.

4. The heating anomaly identification method based on instantaneous flow rate deviation per unit area as described in claim 1, characterized in that, In the step of obtaining the relative deviation of instantaneous flow per unit area of ​​adjacent users on the same floor, the adjacent users are a preset number of users who are pre-selected, are on the same floor as the target user, are on the same heating riser, and whose heating area deviation does not exceed a preset ratio. The step of obtaining the relative deviation of instantaneous flow per unit area of ​​adjacent users on the same floor when the absolute value of the cumulative statistical deviation exceeds the preset abnormal threshold is as follows: when the absolute value of the cumulative statistical deviation of a preset number of consecutive sampling periods exceeds the first preset threshold, cross-validation of neighboring households is triggered to obtain the relative deviation of instantaneous flow per unit area of ​​adjacent users on the same floor.

5. The heating anomaly identification method based on instantaneous flow rate deviation per unit area as described in claim 1, characterized in that, The step of identifying abnormal states and outputting corresponding abnormal levels based on the comparison result of the relative deviation of instantaneous flow per unit area of ​​adjacent users with the preset abnormal threshold includes: If the absolute value of the cumulative statistical deviation of the target user is greater than or equal to the first preset threshold, and the absolute value of the cumulative statistical deviation of its adjacent users in the same layer is less than or equal to the second preset threshold, then it is determined to be a single-household local anomaly, and the first anomaly level is output. If the absolute value of the cumulative deviation of the target user is greater than or equal to the first preset threshold, and the absolute value of the cumulative deviation of the adjacent users on the same floor exceeding a preset proportion is greater than or equal to the third preset threshold, then it is determined to be a building-level global hydraulic imbalance, and the second abnormal level is output.

6. The heating anomaly identification method based on instantaneous flow rate deviation per unit area as described in claim 5, characterized in that, It also includes the following steps: Construct an anomaly severity index The calculation formula is as follows: In the formula, The deviation is calculated cumulatively for the target users. The first preset threshold, The first nonlinear weighting coefficient, The number of adjacent users, For the first The deviation after cumulative statistics of adjacent users The second preset threshold or the third preset threshold, The second nonlinear weighting coefficient, The standard deviation of the cumulative deviations of all adjacent users. This is the deviation dispersion adjustment coefficient; According to the severity index of the anomaly The preset value range in which the first or second abnormality level is located is used to refine the first or second abnormality level into multiple sub-levels, and corresponding sub-level warning signals are output.

7. The heating anomaly identification method based on instantaneous flow rate deviation per unit area as described in claim 5, characterized in that, It also includes the following steps: Calculate the spatial dispersion index of the deviation between the target user and all its neighboring users. and homogeneity coefficient : In the formula, The number of adjacent users, For the first The deviation after cumulative statistics of adjacent users The mean of the absolute values ​​of the deviations between adjacent users. Use extremely small positive numbers to prevent division by zero errors; When the homogeneity coefficient Within a preset global imbalance range and the spatial dispersion index When the deviation is less than the preset upper limit of dispersion, it is determined that the deviation of adjacent user groups is in the same direction, the fault type is building-level global hydraulic imbalance, and the second abnormal level is upgraded to the first-level warning priority. When the homogeneity coefficient The spatial dispersion index is greater than or equal to the first homogeneity threshold and the spatial dispersion index When the dispersion is less than the preset upper limit, the target user is determined to be an isolated anomaly, the fault type is a single-household local anomaly, and the first anomaly level is maintained. When the homogeneity coefficient The spatial dispersion index is located in the preset middle range. When the deviation exceeds the preset lower limit of dispersion, it is determined that the deviations of adjacent user groups are independent of each other, and the fault type is multiple isolated local anomalies superimposed, each maintaining its own first anomaly level output.

8. The heating anomaly identification method based on instantaneous flow rate deviation per unit area as described in claim 1, characterized in that, It also includes the step of: collecting the supply and return water temperature difference of the user's target heating facility or building as a multi-dimensional feature to help verify the abnormal state.

9. The heating anomaly identification method based on instantaneous flow rate deviation per unit area as described in claim 8, characterized in that, The step of identifying abnormal states and outputting corresponding abnormal levels based on the comparison result of the relative deviation of instantaneous flow per unit area of ​​adjacent users with the preset abnormal threshold further includes: Obtain the relative deviation sequence of instantaneous flow per unit area for the target user within the time window. and the corresponding supply and return water temperature difference sequence ; Calculate the time coupling coefficient between the deviation sequence and the supply and return water temperature difference sequence. : In the formula, The arithmetic mean of the supply and return water temperature difference within the time window; Based on the absolute value of the timing coupling coefficient The first preset threshold, the second preset threshold, and the third preset threshold are dynamically adjusted to obtain the adjusted dynamic threshold. The first preset threshold, the second preset threshold, and the third preset threshold are replaced with the corrected dynamic threshold, and the anomaly determination and level output are re-executed.

10. A heating anomaly identification device based on instantaneous flow rate deviation per unit area, characterized in that, include: The data acquisition module is used to collect the real-time heating flow rate, rated design heating flow rate, and corresponding heating building area of ​​the target heating facilities of users in the heating system. The normalized deviation calculation module is used to calculate the relative deviation of instantaneous flow per unit area based on the real-time heating flow, rated design heating flow, and heating building area. The sliding window cumulative filtering module is used to start a time window of preset duration and accumulate and statistically analyze the relative deviations of instantaneous flow rates per unit area within the time window in order to filter out instantaneous fluctuations. The neighboring household data acquisition module is used to acquire the relative deviation of instantaneous flow per unit area of ​​adjacent users on the same floor when the absolute value of the cumulative statistical deviation exceeds the preset abnormal threshold. The anomaly detection and graded output module is used to identify anomalies and output corresponding anomaly levels based on the comparison results between the relative deviation of the instantaneous flow per unit area of ​​the adjacent users and the preset anomaly threshold. The formula for calculating the relative deviation of instantaneous flow rate per unit area is as follows: In the formula, The relative deviation of instantaneous flow rate per unit area. Real-time heat flow rate for the user's target heating facility. The rated design heating flow rate for the user's target heating facility.

Images