Circulating water heat conduction capability abnormity detection method and detection system of heating equipment

Abstract

The invention relates to a circulating water heat conduction capability abnormity detection method and system of heating equipment, and relates to the field of heating equipment, and the method comprises the steps: obtaining a real-time room temperature sequence reported by room temperature monitoring equipment through the Internet of Things; counting a room temperature rising slope sequence and a room temperature falling slope sequence corresponding to the real-time room temperature sequence; determining a room temperature rise rate, a room temperature drop rate, a room temperature rise rate standard fraction and a room temperature drop rate standard fraction; inputting the room temperature rise rate, the room temperature drop rate, the standard score of the room temperature rise rate, the standard score of the room temperature drop rate, the room temperature rise slope sequence and the room temperature drop slope sequence into a pre-trained Wideamp; according to the Deep neural network model, the circulating water heat conduction capacity detection result and the target adjusting power of the to-be-monitored heating equipment are obtained. By adopting the method, the circulating water heat conduction capability of the heating equipment can be accurately evaluated.

Description

Technical Field

[0001] The present application relates to the field of heating equipment, and in particular to a method and system for detecting abnormal thermal conductivity of circulating water in heating equipment. Background Art

[0002] At present, circulating water plays a vital role in heating equipment such as wall-mounted boilers. The thermal conductivity of circulating water affects the effect of circulating water in heating equipment to transfer heat to the room. If the thermal conductivity of circulating water is low, the user's heating needs will be difficult to meet.

[0003] Insufficient thermal conductivity of the circulating water in the heating equipment often means insufficient circulating water volume or abnormal water flow. Insufficient circulating water volume will cause the outlet water temperature of the heating equipment to quickly set the temperature and stop working. The heating equipment will start and stop more frequently, affecting the service life of the heating equipment.

[0004] Therefore, there is an urgent need for a method for detecting abnormal thermal conductivity of circulating water in heating equipment. Summary of the invention

[0005] Based on this, it is necessary to provide a method and system for detecting abnormal thermal conductivity of circulating water in a heating equipment in response to the above-mentioned technical problems.

[0006] In a first aspect, a method for detecting abnormal thermal conductivity of circulating water in a heating device is provided, the method being applied to a cloud central server, the method comprising:

[0007] Obtain the real-time room temperature sequence reported by the room temperature monitoring device through the Internet of Things, and the real-time power reported by the heating equipment through the Internet of Things;

[0008] Counting a room temperature rising slope sequence and a room temperature falling slope sequence corresponding to the real-time room temperature sequence;

[0009] Determine the room temperature rise rate, room temperature drop rate, room temperature rise rate standard score and room temperature drop rate standard score;

[0010] The room temperature rise rate, the room temperature drop rate, the room temperature rise rate standard score, the room temperature drop rate standard score, the real-time power, the room temperature rise slope sequence and the room temperature drop slope sequence are input into a pre-trained Wide&Deep neural network model to obtain the detection result of the circulating water thermal conductivity of the heating equipment to be monitored.

[0011] As an optional implementation manner, the counting of the room temperature rising slope sequence and the room temperature falling slope sequence corresponding to the real-time room temperature sequence includes:

[0012] identifying room temperature peaks and room temperature troughs in the real-time room temperature sequence;

[0013] The room temperature rising slope from the room temperature wave trough to the next room temperature wave peak is counted as the room temperature rising slope sequence;

[0014] The room temperature decreasing slope from the room temperature peak to the next room temperature trough is counted as the room temperature decreasing slope sequence.

[0015] As an optional implementation manner, the determining of the room temperature rise rate, the room temperature drop rate, the room temperature rise rate standard score and the room temperature drop rate standard score comprises:

[0016] Determine the mean value of the slope of the room temperature rising slope sequence as the room temperature rising rate;

[0017] Determine the mean value of the slope of the room temperature decrease slope sequence as the room temperature decrease rate;

[0018] Calculating a standard score of the room temperature rise rate in a pre-stored room temperature rise rate sample population as the room temperature rise rate standard score;

[0019] The standard score of the room temperature drop rate in the pre-stored room temperature drop rate sample population is calculated as the room temperature drop rate standard score.

[0020] As an optional implementation, the training method of the Wide&Deep neural network model includes:

[0021] For each user, obtain the historical room temperature sequence of the preset time period reported by the room temperature monitoring device of the user and the historical power reported by the heating equipment in the same time period;

[0022] Counting a historical room temperature rising slope sequence and a historical room temperature falling slope sequence corresponding to the historical room temperature sequence;

[0023] Based on the historical room temperature rising slope sequence and the historical room temperature falling slope sequence, determining the historical room temperature rising rate, the historical room temperature falling rate, the historical room temperature rising rate standard score and the historical room temperature falling rate standard score;

[0024] Determine the evaluation result of the thermal conductivity of circulating water of the user's heating equipment according to the historical room temperature rise rate standard score, the historical room temperature drop rate standard score, the preset first standard score threshold and the preset second standard score threshold;

[0025] The historical room temperature rise rate, the historical room temperature decrease rate, the historical room temperature rise rate standard score, the historical room temperature decrease rate standard score and the historical power are field-joined and placed in the Wide part of the Wide&Deep neural network model;

[0026] Perform field splicing on the historical room temperature rising slope sequence and the historical room temperature falling slope sequence, and put them into the Deep part of the Wide&Deep neural network model;

[0027] Determining the circulating water thermal conductivity evaluation result as the training data label of the user;

[0028] After the training is completed, the Wide&Deep neural network model is saved and deployed to the cloud.

[0029] As an optional implementation manner, the statistically analyzing a historical room temperature rising slope sequence and a historical room temperature falling slope sequence corresponding to the historical room temperature sequence includes:

[0030] Identifying room temperature peaks and room temperature troughs in the historical room temperature sequence;

[0031] The room temperature rise slope from the room temperature trough to the next room temperature peak is counted as the historical room temperature rise slope sequence;

[0032] The room temperature drop slope from the room temperature peak to the next room temperature trough is counted as the historical room temperature drop slope sequence.

[0033] As an optional implementation manner, determining the historical room temperature rise rate, the historical room temperature drop rate, the historical room temperature rise rate standard score, and the historical room temperature drop rate standard score based on the historical room temperature rise slope sequence and the historical room temperature drop slope sequence includes:

[0034] Determine the slope mean of the historical room temperature rise slope sequence as the historical room temperature rise rate;

[0035] Determine the slope mean of the historical room temperature decrease slope sequence as the historical room temperature decrease rate;

[0036] Calculating a standard score of the historical room temperature rise rate in a pre-stored room temperature rise rate sample population as the standard score of the historical room temperature rise rate;

[0037] The standard score of the historical room temperature drop rate in the pre-stored room temperature drop rate sample population is calculated as the standard score of the historical room temperature drop rate.

[0038] As an optional implementation manner, determining the evaluation result of the thermal conductivity of circulating water of the user's heating equipment according to the historical room temperature rise rate standard score, the historical room temperature drop rate standard score, the preset first standard score threshold and the preset second standard score threshold includes:

[0039] If the historical room temperature rise rate standard score is greater than the first standard score threshold, it is determined that the user's heating equipment has insufficient circulating water under the historical power;

[0040] If the historical room temperature drop rate standard score is greater than the second standard score threshold, it is determined that the heating water heat dissipation rate of the user's heating equipment is insufficient;

[0041] If the user's heating equipment has insufficient circulating water and / or the heating water heat dissipation rate is insufficient, then the evaluation result of the thermal conductivity of the user's heating equipment's circulating water is abnormal, otherwise it is normal.

[0042] As an optional implementation, the circulating water thermal conductivity detection result includes normal and abnormal.

[0043] As an optional implementation, the method further includes:

[0044] When the detection result of the circulating water thermal conductivity is abnormal, the target adjustment power is sent to the user, so that the user can adjust the power of the heating equipment according to the target adjustment power.

[0045] In a second aspect, a detection system is provided, the system comprising a room temperature monitoring device, a heating device and a cloud central server using the method described in any one of the first aspects.

[0046] The present application provides a method and system for detecting abnormal thermal conductivity of circulating water of a heating equipment. The technical solution provided by the embodiment of the present application at least brings the following beneficial effects: obtaining the real-time room temperature sequence reported by the room temperature monitoring device through the Internet of Things; counting the room temperature rising slope sequence and the room temperature falling slope sequence corresponding to the real-time room temperature sequence; determining the room temperature rising rate, the room temperature falling rate, the room temperature rising rate standard score and the room temperature falling rate standard score; inputting the room temperature rising rate, the room temperature falling rate, the room temperature rising rate standard score, the room temperature falling rate standard score, the room temperature rising slope sequence and the room temperature falling slope sequence into a pre-trained Wide & Deep neural network model to obtain the detection result of the thermal conductivity of circulating water of the heating equipment to be monitored and the target adjustment power. The present application uses real-time room temperature data to automatically count the room temperature rising rate and falling rate, and combines the pre-trained Wide & Deep neural network model to accurately evaluate the thermal conductivity of circulating water of the heating equipment. This avoids the errors and delays caused by manual diagnosis, and can find system problems more quickly and accurately. In addition, the embodiment of the present application not only evaluates the room temperature change trend, but also converts these trends into standard scores and compares them with the overall historical samples. This statistically based evaluation method can extract more universal and representative patterns from the data, ensuring that abnormalities in equipment performance can be identified early.

[0047] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0049] Figure 1 A schematic diagram of the structure of a detection system provided in an embodiment of the present application;

[0050] Figure 2 A flow chart of a method for detecting abnormal thermal conductivity of circulating water in a heating device provided in an embodiment of the present application;

[0051] Figure 3 A flow chart of another method for detecting abnormal thermal conductivity of circulating water in a heating device provided in an embodiment of the present application;

[0052] Figure 4 A flow chart of another method for detecting abnormal thermal conductivity of circulating water in a heating device provided in an embodiment of the present application;

[0053] Figure 5 A flowchart of a training method for a Wide&Deep neural network model provided in an embodiment of the present application;

[0054] Figure 6 A flowchart of another Wide&Deep neural network model training method provided in an embodiment of the present application;

[0055] Figure 7 A flowchart of another Wide & Deep neural network model training method provided in an embodiment of the present application;

[0056] Figure 8 A flowchart of another Wide&Deep neural network model training method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0058] The method for detecting abnormal thermal conductivity of circulating water in a heating device provided in the embodiment of the present application can be applied to a cloud central server. Figure 1 As shown, the heating device 110 can report the real-time power to the cloud central server 140 through the Internet of Things 130, and the room temperature monitoring device 120 can report the real-time room temperature to the cloud central server 140 through the Internet of Things 130. The cloud central server 140 is deployed with a pre-trained Wide & Deep neural network model, which can diagnose whether the thermal conductivity of the circulating water is abnormal based on the real-time room temperature reported by the heating device 110, and give a target adjustment power, so that the user or the heating device 110 can adjust the power according to the target adjustment power.

[0059] The following will describe in detail a method for detecting abnormal thermal conductivity of circulating water in a heating device provided in an embodiment of the present application in combination with a specific implementation manner. Figure 2 A flow chart of a method for detecting abnormal thermal conductivity of circulating water in a heating device provided in an embodiment of the present application, the method is applied to a cloud central server, such as Figure 2 As shown, the specific steps are as follows:

[0060] Step 201, obtaining the real-time room temperature sequence reported by the room temperature monitoring device through the Internet of Things, and the real-time power reported by the heating equipment through the Internet of Things.

[0061] In practice, obtaining real-time room temperature series through room temperature monitoring devices is achieved through IoT technology. Each room temperature monitoring device periodically reports the indoor temperature to the server to form time series data. The real-time room temperature series reflects the trend of the user's room temperature over time. For example: a residence is equipped with multiple IoT temperature sensors, which report the indoor temperature every 10 minutes. The cloud central server records the room temperature data for 24 consecutive hours, forming a room temperature time series containing 144 data points. The heating equipment can stream the real-time power according to the preset reporting time interval.

[0062] Step 202 , counting the room temperature rising slope sequence and the room temperature falling slope sequence corresponding to the real-time room temperature sequence.

[0063] In the implementation, the acquired real-time room temperature sequence is statistically processed to calculate the slope of the room temperature change. The slope reflects the rate of rise or fall of the room temperature in a specific time period. By calculating the ratio of the temperature change in each time interval to the time interval, the room temperature rise slope and fall slope sequence can be obtained, thereby more finely characterizing the dynamic process of temperature change. For example: for the above 24-hour room temperature sequence, the room temperature change slope every 10 minutes is statistically obtained. Assuming that the temperature rises from 20°C to 22°C within a certain period of time, and this period of time is 30 minutes, the corresponding room temperature rise slope is (22-20) / 30=0.067°C / min.

[0064] As an optional implementation, Figure 3 A flowchart of another method for detecting abnormal thermal conductivity of circulating water in a heating device provided in an embodiment of the present application is shown in FIG. Figure 3 As shown, the specific steps of counting the room temperature rising slope sequence and the room temperature falling slope sequence corresponding to the real-time room temperature sequence in step 202 are as follows:

[0065] Step 301, identifying room temperature peaks and room temperature troughs in a real-time room temperature sequence.

[0066] In practice, a peak is the position where the room temperature reaches a local maximum at a certain moment, and a trough is the position where the room temperature reaches a local minimum at a certain moment. These peaks and troughs can be detected by processing the room temperature sequence through an algorithm (such as a sliding window or derivative method). Assume that the real-time room temperature sequence is: 22°C, 23°C, 24°C, 23.5°C, 23°C, and 22.5°C. In this sequence, 24°C is the peak and 22°C is the trough. By analyzing these local extreme points, the fluctuation period of the room temperature can be identified.

[0067] Step 302: Count the room temperature rising slope from the room temperature trough to the next room temperature peak as a room temperature rising slope sequence.

[0068] In practice, after identifying the trough and peak, the temperature rise process from each trough to the next peak is slope calculated to obtain a room temperature rise slope sequence. The specific calculation method is: room temperature rise slope = (peak temperature - trough temperature) / (time interval between peak and trough). This can quantify the speed of each temperature rise. Assume that the trough temperature is 22°C, the peak temperature is 24°C, and the time interval from trough to peak is 30 minutes. The corresponding room temperature rise slope is (24-22) / 30 = 0.067°C / min. This result will be recorded in the room temperature rise slope sequence.

[0069] Step 303: Count the room temperature drop slope from the room temperature peak to the next room temperature trough into a room temperature drop slope sequence.

[0070] In implementation, similar to the room temperature rising slope sequence, the temperature drop process from each peak to the next trough is slope calculated to obtain the room temperature drop slope sequence. The specific calculation method is: room temperature drop slope = (room temperature at the trough point - room temperature at the peak point) / (time interval between the peak and the trough) to quantify the speed of each room temperature drop. Assuming that the room temperature at the peak point is 24°C, the room temperature at the trough point is 22.5°C, and the time interval from the peak to the trough is 20 minutes, the corresponding room temperature drop slope is (22.5-24) / 20 = -0.075°C / min. This result will be recorded in the room temperature drop slope sequence.

[0071] Step 203, determining the room temperature rise rate, room temperature drop rate, room temperature rise rate standard score and room temperature drop rate standard score.

[0072] In practice, in order to quantify the characteristics of room temperature changes, in addition to calculating the room temperature rise and fall rates, it is also necessary to calculate its standard score (z-score), that is, to compare each slope with the overall mean. The standardized score is convenient for comparison between different data sets. By calculating the standard score of each data point, its degree of deviation from the overall temperature change trend can be determined.

[0073] As an optional implementation, Figure 4 A flow chart of another method for detecting abnormal thermal conductivity of circulating water in a heating device provided in an embodiment of the present application is as follows: Figure 4 As shown, the specific steps of determining the room temperature rise rate, the room temperature drop rate, the room temperature rise rate standard score and the room temperature drop rate standard score in step 203 are as follows:

[0074] Step 401: determine the slope mean of the room temperature rising slope sequence as the room temperature rising rate.

[0075] In implementation, the room temperature rise rate is determined by calculating the mean of the room temperature rise slope sequence. The rising slope sequence contains multiple slopes from trough to peak. By taking the average of these slopes, the overall room temperature rise rate over a period of time can be reflected. This average value is the room temperature rise rate, which indicates the overall trend of room temperature rise during the monitoring period. During a 24-hour monitoring period, the room temperature rise slope sequence collected is [0.05, 0.06, 0.07] ℃ / min. By calculating its mean, the room temperature rise rate is (0.05+0.06+0.07) / 3=0.06℃ / min.

[0076] Step 402: determine the mean value of the slope of the room temperature decrease slope sequence as the room temperature decrease rate.

[0077] In practice, the room temperature drop rate is determined by calculating the mean of the room temperature drop slope sequence. The drop slope sequence includes multiple slopes from peak to trough. By calculating its average value, the overall drop rate of the room temperature over a period of time can be obtained. The average value is the room temperature drop rate, which is used to indicate the overall trend of the room temperature drop. In the same 24-hour monitoring period, the room temperature drop slope sequence collected is [-0.04, -0.05, -0.06] ℃ / min. By calculating its mean, the room temperature drop rate is (-0.04+-0.05+-0.06) / 3=-0.05℃ / min.

[0078] Step 403, calculating the standard score of the room temperature rise rate in the pre-stored room temperature rise rate sample population as the room temperature rise rate standard score.

[0079] In implementation, in order to compare the currently calculated room temperature rise rate with historical data, its standard score (z-score) needs to be calculated. The standard score determines its relative position by comparing the current room temperature rise rate with the mean and standard deviation of the pre-stored historical sample population. The specific formula is: (current rise rate-historical mean) / standard deviation. This score can reflect the relative extreme degree of the current rise rate in the historical data. Assume that the mean of the pre-stored room temperature rise rate sample population is 0.05℃ / min and the standard deviation is 0.01℃ / min. If the room temperature rise rate calculated in the current monitoring period is 0.06℃ / min, the standard score is (0.06-0.05) / 0.01=1. This means that the current room temperature rise rate is 1 standard deviation higher than the historical average.

[0080] Step 404, calculating the standard score of the room temperature drop rate in the pre-stored room temperature drop rate sample population as the room temperature drop rate standard score.

[0081] In implementation, similar to the standard score for the room temperature rise rate, the standard score for the room temperature decrease rate is calculated by comparing the current decrease rate with the pre-stored historical sample population. The standard score formula is: (current decrease rate-historical mean) / standard deviation. This helps to determine whether the current room temperature decrease rate is abnormal or within the normal range. Assume that the mean of the pre-stored room temperature decrease rate sample population is -0.04℃ / min and the standard deviation is 0.01℃ / min. If the room temperature decrease rate calculated during the current monitoring period is -0.05℃ / min, the standard score is (-0.05-(-0.04)) / 0.01=-1. This indicates that the current room temperature decrease rate is 1 standard deviation lower than the historical average.

[0082] Step 204, input the room temperature rise rate, room temperature drop rate, room temperature rise rate standard score, room temperature drop rate standard score, real-time power, room temperature rise slope sequence and room temperature drop slope sequence into the pre-trained Wide & Deep neural network model to obtain the circulating water thermal conductivity detection result of the heating equipment to be monitored.

[0083] In implementation, the aforementioned room temperature rise rate, fall rate and their standard scores, real-time power, and slope sequence are input into the pre-trained Wide&Deep neural network model. This model combines the advantages of generalized linear models and deep neural networks, and can capture both linear relationships and complex nonlinear relationships. Through the training and reasoning of this model, the circulating water thermal conductivity and adjustment power of the heating equipment can be predicted. The output results of the model can be used to adjust the operation of the heating system to ensure room temperature stability and optimal energy efficiency. In actual operation, the model inputs these real-time data, outputs the circulating water thermal conductivity test result as "normal", and can also output the recommended target adjustment power.

[0084] As an optional implementation manner, the circulating water thermal conductivity detection result includes normal and abnormal.

[0085] As an optional implementation, when the detection result of the circulating water thermal conductivity is abnormal, the target adjustment power is sent to the user, so that the user can adjust the power of the heating equipment according to the target adjustment power.

[0086] In practice, when the test result of circulating water thermal conductivity is abnormal, it means that the thermal conductivity efficiency of the heating equipment is not within the normal range, and there may be system failure or low operating efficiency. To this end, the Wide&Deep neural network model can output an optimized target adjustment power and send it to the user. The target adjustment power is the optimal heating power predicted by the neural network model based on real-time room temperature changes and equipment status, which aims to ensure that the indoor temperature is maintained within the set range while improving energy efficiency. Users can adjust the power settings of the heating equipment according to the suggestions provided by the system to restore or optimize the heating effect.

[0087] As an optional implementation, Figure 5 A flowchart of a training method for a Wide & Deep neural network model provided in an embodiment of the present application, such as Figure 5 As shown, the specific steps are as follows:

[0088] Step 501 : for each user, obtain the historical room temperature sequence of a preset time period reported by the room temperature monitoring device of the user and the historical power reported by the heating device in the same time period.

[0089] In implementation, in order to train the Wide&Deep neural network model, it is first necessary to obtain the user's historical data. Specifically, the historical room temperature sequence within a certain period of time is obtained from the user's room temperature monitoring device, and the historical power data of the corresponding period is obtained from the heating equipment. By pairing these two parts of data, the relationship between the operation effect of the heating system and the heating power can be provided to the model. A user's room temperature monitoring device reported temperature data once an hour in the past 14 days, and the heating equipment recorded its operating power in the same period. In this way, a 14-day historical room temperature sequence and the corresponding historical power data can be obtained.

[0090] Step 502: Count the historical room temperature rising slope sequence and the historical room temperature falling slope sequence corresponding to the historical room temperature sequence.

[0091] In practice, after obtaining historical room temperature data, these room temperature sequences are processed to calculate the historical room temperature rising slope sequence and historical falling slope sequence. This is similar to the previous real-time monitoring, which determines the rising and falling speeds of the room temperature by calculating the slopes from each trough to the peak and from the peak to the trough.

[0092] As an optional implementation, Figure 6 A flowchart of another Wide & Deep neural network model training method provided in an embodiment of the present application, such as Figure 6 As shown, the specific steps of calculating the historical room temperature rising slope sequence and the historical room temperature falling slope sequence corresponding to the historical room temperature sequence in step 502 are as follows:

[0093] Step 601, identifying room temperature peaks and room temperature troughs in a historical room temperature sequence.

[0094] In practice, identifying the room temperature peaks and troughs in the historical room temperature series is mainly done by analyzing the changing trends of the room temperature at different time points. The peak is the moment when the temperature reaches a local maximum point, and the trough is the moment when the temperature reaches a local minimum point. Time series analysis methods, such as sliding windows or derivative methods, can be used to detect these local extreme points.

[0095] Step 602: Count the room temperature rising slope from the room temperature trough to the next room temperature peak as a historical room temperature rising slope sequence.

[0096] In practice, after identifying the troughs and peaks in the historical room temperature, the temperature change from each trough to the next peak is calculated as the rising slope. The specific formula is: rising slope = (peak temperature - trough temperature) / (time interval between peak and trough). This slope sequence records the rate of temperature rise for subsequent analysis. For example: assuming the trough temperature is 20°C, the peak temperature is 22°C, and the time interval from trough to peak is 60 minutes, the room temperature rising slope is (22-20) / 60 = 0.033°C / min. This rising slope will be recorded in the historical room temperature rising slope sequence.

[0097] Step 603: Count the room temperature drop slope from the room temperature peak to the next room temperature trough as a historical room temperature drop slope sequence.

[0098] In practice, similarly, the temperature change from each peak to the next trough is calculated as the descending slope. The specific formula is: descending slope = (trough temperature - peak temperature) / (time interval between peak and trough). This slope sequence records the rate of temperature decline and is used to evaluate the thermal conductivity of heating equipment. Assuming the peak temperature is 22°C, the trough temperature is 20.5°C, and the time interval from peak to trough is 40 minutes, the room temperature decline slope is (20.5-22) / 40 = -0.038°C / min. This decline slope will be recorded in the historical room temperature decline slope sequence.

[0099] Step 503, based on the historical room temperature rising slope sequence and the historical room temperature falling slope sequence, determine the historical room temperature rising rate, the historical room temperature falling rate, the historical room temperature rising rate standard score and the historical room temperature falling rate standard score.

[0100] In implementation, based on the slope sequence obtained in step 502, the average values ​​of the room temperature rise rate and the room temperature fall rate of these historical data are calculated and further standardized by the standard score. The standard score is used to evaluate the degree of deviation of the heating effect of the current user relative to the overall situation, which helps to reveal whether there are abnormal conditions. The average value of the historical room temperature rise slope is 0.06℃ / min, and the standard deviation is 0.01℃ / min. Through the standard score formula, the standard score of the historical room temperature rise rate calculated is 2, indicating that the user's room temperature rises faster.

[0101] As an optional implementation, Figure 7 A flowchart of another method for training a Wide & Deep neural network model provided in an embodiment of the present application is shown in FIG. Figure 7 As shown, in step 503, based on the historical room temperature rising slope sequence and the historical room temperature falling slope sequence, the specific steps of determining the historical room temperature rising rate, the historical room temperature falling rate, the historical room temperature rising rate standard score and the historical room temperature falling rate standard score are as follows:

[0102] Step 701: determine the slope mean of the historical room temperature rise slope sequence as the historical room temperature rise rate.

[0103] In implementation, the historical room temperature rise rate is determined by calculating the mean of the historical room temperature rise slope sequence. The rising slope sequence records the rising speed from multiple troughs to peaks. Taking the average of these slopes can obtain the overall rising trend over a period of time. Through this historical rise rate, the overall heating efficiency of the heating equipment can be evaluated. Assuming that the historical room temperature rise slope sequence of a user is [0.03, 0.04, 0.05] ℃ / min, the mean of its slope is calculated as (0.03+0.04+0.05) / 3=0.04℃ / min. This 0.04℃ / min is the historical room temperature rise rate of the user.

[0104] Step 702: Determine the slope mean of the historical room temperature drop slope sequence as the historical room temperature drop rate.

[0105] In implementation, the historical room temperature drop rate is determined by calculating the mean of the historical room temperature drop slope sequence. The drop slope sequence records the temperature drop rate from the peak to the trough. Taking the average of these slopes can reflect the cooling efficiency or heat dissipation of the equipment. Assuming that the historical room temperature drop slope sequence is [-0.02, -0.03, -0.04] ℃ / min, the slope mean is calculated as (-0.02+-0.03+-0.04) / 3=-0.03℃ / min. This -0.03℃ / min is the historical room temperature drop rate.

[0106] Step 703, calculating the standard score of the historical room temperature rise rate in the pre-stored room temperature rise rate sample population as the historical room temperature rise rate standard score.

[0107] In implementation, in order to further determine whether the historical room temperature rise rate is normal, it is necessary to calculate its standard score in the pre-stored sample population. The standard score determines its relative position in the sample by comparing the historical rise rate with the mean and standard deviation of the sample population. The formula is: (historical rise rate-sample population mean) / sample population standard deviation. This score is used to determine whether the heating efficiency of the current device meets expectations. Assuming that the mean of the historical room temperature rise rate of the sample population is 0.035℃ / min and the standard deviation is 0.01℃ / min, if the historical rise rate of the current user is 0.04℃ / min, then its standard score is (0.04-0.035) / 0.01=0.5. This shows that the user's rise rate is slightly higher than the average of the overall sample.

[0108] Step 704, calculating the standard score of the historical room temperature drop rate in the pre-stored room temperature drop rate sample population as the standard score of the historical room temperature drop rate.

[0109] In implementation, the standard score of the historical room temperature drop rate is calculated in a similar way, comparing the historical drop rate with the mean and standard deviation of the sample population. This standard score is used to evaluate the difference in the heat dissipation performance of the heating equipment with the overall sample. Assuming that the mean of the historical room temperature drop rate of the sample population is -0.03℃ / min and the standard deviation is 0.01℃ / min, if the historical drop rate of the current user is -0.04℃ / min, then its standard score is (-0.04-(-0.03)) / 0.01=-1. This indicates that the user's drop rate is lower than the overall sample and the heat dissipation speed is faster.

[0110] Step 504, determining the evaluation result of the thermal conductivity of circulating water of the user's heating equipment according to the historical room temperature rise rate standard score, the historical room temperature drop rate standard score, the preset first standard score threshold and the preset second standard score threshold.

[0111] In implementation, the calculated historical room temperature rise rate standard score and drop rate standard score are compared with the preset first and second standard score thresholds. If a user's standard score exceeds or falls below these thresholds, the system will evaluate the user's heating equipment as having abnormal circulating water thermal conductivity, otherwise it will be normal. The preset first standard score threshold is 1.5, and the second standard score threshold is -1.5. If the user's historical room temperature rise rate standard score is 2, which exceeds the first standard score threshold, the circulating water thermal conductivity of its heating equipment is evaluated as "high", and attention needs to be paid to the energy efficiency of the equipment.

[0112] As an optional implementation, Figure 8 A flowchart of another method for training a Wide & Deep neural network model provided in an embodiment of the present application is shown in FIG. Figure 8 As shown, in step 504, the specific steps of determining the evaluation result of the circulating water thermal conductivity of the user's heating equipment according to the historical room temperature rise rate standard score, the historical room temperature drop rate standard score, the preset first standard score threshold and the preset second standard score threshold are as follows:

[0113] Step 801: If the historical room temperature rise rate standard score is greater than the first standard score threshold, it is determined that the user's heating equipment has insufficient circulating water under the historical power.

[0114] In implementation, when the standard score of the user's historical room temperature rise rate is greater than the preset first standard score threshold, it means that the temperature of the heating equipment rises too quickly under the historical power. This usually indicates that the flow rate of the circulating water is insufficient, and the heat generated by the heating cannot be fully absorbed and transferred, resulting in a rapid rise in the room temperature. Therefore, in this case, it can be determined that the user's heating equipment has a problem of insufficient circulating water. Assuming that the first standard score threshold is 3, if a user's historical room temperature rise rate standard score is 3.5, it indicates that the user's room temperature rises quickly. This may be due to insufficient circulating water flow in the heating system, which cannot effectively disperse the generated heat, resulting in excessive temperature rise. Therefore, the system determines that the user's heating equipment has a problem of insufficient circulating water.

[0115] Step 802: If the standard score of the historical room temperature drop rate is greater than the second standard score threshold, it is determined that the heat dissipation rate of the heating water of the user's heating equipment is insufficient.

[0116] In implementation, if the user's historical room temperature drop rate standard score is greater than the preset second standard score threshold, it indicates that the temperature of the heating equipment drops slowly, which means that the heat dissipation rate of the heating water is insufficient. Insufficient heat dissipation rate may be due to the inefficiency of the equipment's heat dissipation components (such as radiators) or the presence of factors that hinder heat dissipation in the system. Assuming that the second standard score threshold is 3.0, if the user's historical room temperature drop rate standard score is 3.2, it indicates that the user's heating water cools down slowly. This may be due to the low efficiency of the heating system radiator, which prevents heat from being dissipated in time. Therefore, the system determines that the user's heating equipment has a problem of insufficient heating water heat dissipation rate.

[0117] Step 803: If the circulating water of the user's heating equipment is insufficient and / or the heat dissipation rate of the heating water is insufficient, the evaluation result of the thermal conductivity of the circulating water of the user's heating equipment is abnormal, otherwise it is normal.

[0118] In implementation, if it is detected that the user's heating equipment has insufficient circulating water and / or insufficient heat dissipation rate of the heating water, an evaluation result of abnormal thermal conductivity of the user's circulating water can be obtained. Conversely, if both indicators are within the normal range, the thermal conductivity of the equipment is judged to be normal. The standard score of the historical room temperature rise rate of a certain user is 2.0 (insufficient circulating water), and the standard score of the drop rate is 0.8 (normal heat dissipation rate), so the system determines that the thermal conductivity of the user's circulating water is abnormal. If both are below the set threshold, the thermal conductivity evaluation result of the user is normal.

[0119] Step 505, concatenate the historical room temperature rise rate, the historical room temperature drop rate, the historical room temperature rise rate standard score, the historical room temperature drop rate standard score and the historical power, and put them into the Wide part of the Wide&Deep neural network model.

[0120] In the implementation, the user's historical room temperature rise rate, drop rate and their corresponding standard scores are spliced ​​with the historical power as the Wide part input of the Wide&Deep neural network model. The Wide part processes linear features and is suitable for factors that directly affect heating efficiency. The user's historical room temperature rise rate of 0.06℃ / min, drop rate of -0.05℃ / min, standard scores of 2 and -1, and historical heating power in the corresponding period are spliced ​​to form the input data of the Wide part.

[0121] Step 506, concatenate the fields of the historical room temperature rising slope sequence and the historical room temperature falling slope sequence, and put them into the Deep part of the Wide & Deep neural network model.

[0122] In the implementation, the historical room temperature rising slope sequence and falling slope sequence are concatenated as the input of the Deep part. The Deep part processes nonlinear features and is suitable for capturing complex time series changes and the operating dynamics of the heating system. The user's room temperature rising slope sequence (such as [0.05, 0.06, 0.07]) and falling slope sequence (such as [-0.04, -0.05, -0.06]) of the past 30 days are concatenated as the input data of the Deep part to train the model to capture the complex pattern of room temperature changes.

[0123] Step 507: Determine the evaluation result of the circulating water thermal conductivity as the training data label of the user.

[0124] In implementation, the thermal conductivity evaluation results of each user's heating equipment circulating water will be used as labels for training data. Through these labels, the model can learn the relationship between room temperature change characteristics and thermal conductivity, thus providing a basis for future inferences. The user's thermal conductivity evaluation results are "normal" or "abnormal", and these results are paired with the corresponding input features as labels for training the model.

[0125] Step 508, after the training is completed, save the Wide&Deep neural network model and deploy it to the cloud.

[0126] In practice, after training is completed, the Wide&Deep neural network model is saved and deployed to the cloud for practical application. The cloud model can make real-time inferences based on new user data and continuously update to improve prediction accuracy. After training is completed, the model is deployed to the cloud, and the heating system can call the model through the cloud to conduct real-time performance evaluation and optimization adjustment suggestions for heating equipment of different users.

[0127] The embodiment of the present application provides a method for detecting abnormal thermal conductivity of circulating water of a heating equipment, which is applied to a cloud central server, and the method includes: obtaining a real-time room temperature sequence reported by a room temperature monitoring device through the Internet of Things. Statistical room temperature rise slope sequence and room temperature drop slope sequence corresponding to the real-time room temperature sequence. Determine the room temperature rise rate, room temperature drop rate, room temperature rise rate standard score and room temperature drop rate standard score. Input the room temperature rise rate, room temperature drop rate, room temperature rise rate standard score, room temperature drop rate standard score, room temperature rise slope sequence and room temperature drop slope sequence into a pre-trained Wide & Deep neural network model to obtain the detection result of the thermal conductivity of circulating water of the heating equipment to be monitored and the target adjustment power. The embodiment of the present application uses real-time room temperature data to automatically count the room temperature rise rate and drop rate, and combines the pre-trained Wide & Deep neural network model to accurately evaluate the thermal conductivity of circulating water of the heating equipment. This avoids the errors and delays caused by manual diagnosis, and can find system problems more quickly and accurately. In addition, the embodiment of the present application not only evaluates the room temperature change trend, but also converts these trends into standard scores and compares them with the overall historical sample. This statistically based evaluation method can extract more universal and representative patterns from the data, ensuring that abnormalities in equipment performance can be identified early.

[0128] It should be understood that although Figures 2 to 8 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figures 2 to 8 At least part of the steps may include multiple steps or multiple stages. These steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.

[0129] It can be understood that the same / similar parts between the various embodiments of the above method in this specification can refer to each other, and each embodiment focuses on the differences from other embodiments. For related points, please refer to the description of other method embodiments.

[0130] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0131] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0132] It should also be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data for analysis, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0133] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0134] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0135] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for detecting abnormal thermal conductivity of circulating water in a heating device, characterized in that: The method is applied to a cloud central server, and the method comprises: Obtain the real-time room temperature sequence reported by the room temperature monitoring device through the Internet of Things, and the real-time power reported by the heating equipment through the Internet of Things; Counting a room temperature rising slope sequence and a room temperature falling slope sequence corresponding to the real-time room temperature sequence; Determine the room temperature rise rate, room temperature drop rate, room temperature rise rate standard score and room temperature drop rate standard score; The room temperature rise rate, the room temperature drop rate, the room temperature rise rate standard score, the room temperature drop rate standard score, the real-time power, the room temperature rise slope sequence and the room temperature drop slope sequence are input into a pre-trained Wide&Deep neural network model to obtain the detection result of the circulating water thermal conductivity of the heating equipment to be monitored.

2. The method according to claim 1, characterized in that The counting of the room temperature rising slope sequence and the room temperature falling slope sequence corresponding to the real-time room temperature sequence includes: identifying room temperature peaks and room temperature troughs in the real-time room temperature sequence; The room temperature rising slope from the room temperature wave trough to the next room temperature wave peak is counted as the room temperature rising slope sequence; The room temperature decreasing slope from the room temperature peak to the next room temperature trough is counted as the room temperature decreasing slope sequence.

3. The method according to claim 1, characterized in that The determining of the room temperature rise rate, the room temperature drop rate, the room temperature rise rate standard score and the room temperature drop rate standard score comprises: Determine the mean value of the slope of the room temperature rising slope sequence as the room temperature rising rate; Determine the mean value of the slope of the room temperature decrease slope sequence as the room temperature decrease rate; Calculating a standard score of the room temperature rise rate in a pre-stored room temperature rise rate sample population as the room temperature rise rate standard score; The standard score of the room temperature drop rate in the pre-stored room temperature drop rate sample population is calculated as the room temperature drop rate standard score.

4. The method according to claim 1, characterized in that: The training method of the Wide&Deep neural network model includes: For each user, obtain the historical room temperature sequence of the preset time period reported by the room temperature monitoring device of the user and the historical power reported by the heating equipment in the same time period; Counting a historical room temperature rising slope sequence and a historical room temperature falling slope sequence corresponding to the historical room temperature sequence; Based on the historical room temperature rising slope sequence and the historical room temperature falling slope sequence, determining the historical room temperature rising rate, the historical room temperature falling rate, the historical room temperature rising rate standard score and the historical room temperature falling rate standard score; Determine the evaluation result of the thermal conductivity of circulating water of the user's heating equipment according to the historical room temperature rise rate standard score, the historical room temperature drop rate standard score, the preset first standard score threshold and the preset second standard score threshold; The historical room temperature rise rate, the historical room temperature decrease rate, the historical room temperature rise rate standard score, the historical room temperature decrease rate standard score and the historical power are field-joined and placed in the Wide part of the Wide&Deep neural network model; Perform field splicing on the historical room temperature rising slope sequence and the historical room temperature falling slope sequence, and put them into the Deep part of the Wide&Deep neural network model; Determining the circulating water thermal conductivity evaluation result as the training data label of the user; After the training is completed, the Wide&Deep neural network model is saved and deployed to the cloud.

5. The method according to claim 4, characterized in that The statistically calculating the historical room temperature rising slope sequence and the historical room temperature falling slope sequence corresponding to the historical room temperature sequence includes: Identifying room temperature peaks and room temperature troughs in the historical room temperature sequence; The room temperature rise slope from the room temperature trough to the next room temperature peak is counted as the historical room temperature rise slope sequence; The room temperature drop slope from the room temperature peak to the next room temperature trough is counted as the historical room temperature drop slope sequence.

6. The method according to claim 4, characterized in that The determining of the historical room temperature rise rate, the historical room temperature drop rate, the historical room temperature rise rate standard score and the historical room temperature drop rate standard score based on the historical room temperature rise slope sequence and the historical room temperature drop slope sequence includes: Determine the slope mean of the historical room temperature rise slope sequence as the historical room temperature rise rate; Determine the slope mean of the historical room temperature decrease slope sequence as the historical room temperature decrease rate; Calculating a standard score of the historical room temperature rise rate in a pre-stored room temperature rise rate sample population as the standard score of the historical room temperature rise rate; The standard score of the historical room temperature drop rate in the pre-stored room temperature drop rate sample population is calculated as the standard score of the historical room temperature drop rate.

7. The method according to claim 4, characterized in that The step of determining the evaluation result of the thermal conductivity of circulating water of the user's heating equipment according to the historical room temperature rise rate standard score, the historical room temperature drop rate standard score, the preset first standard score threshold and the preset second standard score threshold comprises: If the historical room temperature rise rate standard score is greater than the first standard score threshold, it is determined that the user's heating equipment has insufficient circulating water under the historical power; If the historical room temperature drop rate standard score is greater than the second standard score threshold, it is determined that the heating water heat dissipation rate of the user's heating equipment is insufficient; If the user's heating equipment has insufficient circulating water and / or the heating water heat dissipation rate is insufficient, then the evaluation result of the thermal conductivity of the user's heating equipment's circulating water is abnormal, otherwise it is normal.

8. The method according to claim 1, characterized in that The circulating water thermal conductivity test results include normal and abnormal.

9. The method according to claim 8, characterized in that The method further comprises: When the detection result of the circulating water thermal conductivity is abnormal, the target adjustment power is sent to the user, so that the user can adjust the power of the heating equipment according to the target adjustment power.

10. A detection system, characterized in that: The system includes room temperature monitoring equipment, heating equipment and a cloud central server to which the method described in any one of claims 1 to 9 is applied.

Images