A centralized heating data analysis system for decentralized users based on the Internet of Things

Abstract

The present invention discloses a centralized heating data analysis system for decentralized users based on the Internet of Things, which relates to the technical field of centralized heating and solves the technical problem of difficultly processing and analyzing the sampled data of the heating system for each region and time stage of the heating system; the data segmentation processing module divides the heating data into different subsets according to the start-up stage, stable operation stage and shutdown stage of the heating system; by extracting features from each subset of the historical heating data, a normal data recognition model for each stage is established; there may be significant differences in the heating characteristics and demands in the start-up stage, stable operation stage and shutdown stage. By separately modeling each stage, these differences can be better captured, thereby improving the prediction accuracy of the model; at the same time, in the stable operation stage, the state of the system is relatively stable and the data noise is less, which helps to improve the ability of the model to recognize the normal state.

Description

Technical Field

[0001] The present invention belongs to the field of central heating, and specifically relates to a central heating data analysis system for decentralized users based on the Internet of Things. Background Art

[0002] With the acceleration of the urbanization process and the improvement of residents' living standards, the central heating system plays an increasingly important role in modern urban heating. Traditional central heating systems usually rely on fixed heat sources and transmission pipelines, and it is difficult to flexibly meet the diverse heating needs of decentralized users. At the same time, with the development of Internet of Things (IoT) technology, intelligent devices that can collect and analyze large amounts of data in real time have gradually become popular, providing new possibilities for optimizing heating management. In the scenario of central heating for decentralized users, user demands vary greatly and environmental factors are complex, which poses challenges to traditional heating regulation methods. Through IoT technology, real-time monitoring and data collection of each user terminal, pipeline, and equipment can be achieved, thereby realizing more accurate and efficient heating management.

[0003] For most central heating data analysis systems for decentralized users, it is difficult to process and analyze the sampled data of the heating system for each region and time stage by uniformly processing and analyzing the data. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; for this purpose, the present invention provides a central heating data analysis system for decentralized users based on the Internet of Things, which is used to solve the technical problem of being difficult to process and analyze the sampled data of the heating system for each region and time stage of the heating system.

[0005] To solve the above problems, the first aspect of the present invention provides a central heating data analysis system for decentralized users based on the Internet of Things, including:

[0006] Intelligent zoning module: Divide the heating system into a heat source monitoring area, a heat transfer monitoring area, and a heat dissipation monitoring area. Set sampling nodes at the heat output positions in the heat source monitoring area. Divide the heat transfer monitoring area into an underground transfer area and a floor transfer area, and set sampling nodes with different densities in the underground transfer area and the floor transfer area. Set terminal data sampling nodes in the heat dissipation monitoring area;

[0007] Data acquisition module: Collect heating data at the sampling nodes in each area of the heating system; among them, collect pipeline temperature, pressure, sound, and flow data in the underground transfer area of the heat transfer monitoring area, and collect pipeline temperature, pressure, sound, video, and flow data in the floor transfer area.

[0008] Data segmentation processing module: Divide the heating data into different subsets according to the start-up stage, stable operation stage, and shutdown stage of the heating system; establish a normal data recognition model for each stage by extracting features from each subset of historical heating data.

[0009] Real-time data comparison module: Compare the real-time data with the normal data recognition model of the corresponding stage, and perform abnormal recognition analysis on the abnormal data.

[0010] Data analysis module: Dynamically evaluate the heating loss coefficient through the data collected by the sampling nodes in the heat source monitoring area, the heat transfer monitoring area, and the heat dissipation monitoring area. At the same time, predict the real-time heating demand of users, and adjust the heating parameters according to the real-time prediction results of the heating demand and the heating loss coefficient.

[0011] As a further solution of the present invention: Divide the heat transfer monitoring area into an underground transfer area and a floor transfer area, and set sampling nodes with different densities in the underground transfer area and the floor transfer area, including the following steps:

[0012] Set the transmission pipeline buried underground in the heat transfer monitoring area as the underground transfer area, and set the transmission pipeline between the above-ground buildings as the floor transfer area;

[0013] For the underground transfer area, set a sampling node every 500 meters to collect pipeline temperature, pressure, sound, and flow data;

[0014] For the floor transfer area, set a sampling node every 10 meters to collect pipeline temperature, pressure, sound, and flow data, and set video sampling nodes at the positions of the water distributor, the water collector, and the floor main valve.

[0015] As a further solution of the present invention: Collect heating data at the sampling nodes in each area of the heating system, including the following steps:

[0016] Collect the pressure, flow, and temperature data of the heat source outlet and return water through the sampling nodes in the heat source monitoring area;

[0017] Collect pipeline temperature, pressure, sound, video, and flow data through the sampling nodes in the heat transfer monitoring area;

[0018] Collect the pressure, flow, and temperature data of the heat dissipation terminal inlet and return water through the sampling nodes in the heat dissipation monitoring area;

[0019] Number the pipelines and sampling nodes in each area, and add data labels to the heating data collected by the sampling nodes according to the sampling pipeline and node numbers, as well as the sampling timestamp.

[0020] As a further solution of the present invention: the data segmentation processing module divides the heating data into different subsets according to the start-up stage, stable operation stage and shutdown stage of the heating system, including the following steps:

[0021] Set the stage from the shutdown state of the heating system to the stage when the temperature detected by the sampling node in the heat dissipation monitoring area reaches a predetermined temperature as the start-up stage;

[0022] Set the stage from when the temperature detected by the sampling node in the heat dissipation monitoring area reaches a predetermined temperature to before the heat source in the heat source monitoring area stops heating as the stable operation stage;

[0023] Set the process from when the heat source in the heat source monitoring area stops heating to when the flow rate in the pipeline in the heat dissipation monitoring area drops to a preset threshold as the shutdown stage;

[0024] According to the settings of the start-up stage, stable operation stage and shutdown stage of the heating system, divide the heating data into different subsets according to the time stamps of the heating data, and partition the data in the subsets according to the data tags in each subset, and divide the data belonging to the same sampling pipeline in the subsets into the same partition.

[0025] As a further solution of the present invention: by extracting features from each subset of historical heating data, establish a normal data recognition model for each stage, including the following steps:

[0026] Obtain historical heating data, label the data when different areas of the heating system are operating normally as normal data, label the data when different areas of the heating system are operating abnormally as abnormal data, and divide the historical heating data into different data subsets according to the acquisition stage of the historical heating data;

[0027] Extract the sound and video data of the pipeline collected by the sampling node in the heat transfer monitoring area from the historical heating data subset;

[0028] Extract the time-domain features of the audio of the pipeline sound data, including: mean value, variance and peak value, convert the time-domain signal of the audio into a frequency-domain representation through Fourier transform, and extract the frequency-domain features, including: frequency width and power spectral density;

[0029] Mark the pictures with abnormalities in the image data of the positions of the water distribution manifold, water collector and floor main valve in the historical data, and train a deep learning model with the marked image data to identify whether the image data of the positions of the water distribution manifold, water collector and floor main valve is abnormal;

[0030] For the video data of the positions of the water distribution manifold, water collector and floor main valve collected by the sampling node in the heat transfer monitoring area, extract one image frame every 50 frames, and input the extracted image frames into the trained deep learning model to identify whether abnormalities occur;

[0031] Replace the sound data of the pipeline collected by the sampling nodes in the heat transfer monitoring area with the sound data of the pipeline in the time-domain feature and frequency-domain feature replacement subsets, and replace the video data in the heat transfer monitoring area with the judgment result of whether the image appears abnormal;

[0032] Train the MLP multi-layer perceptron respectively through the subset data of the start-up stage, stable operation stage and shutdown stage of the heating system, and identify the subset data of the start-up stage, stable operation stage and shutdown stage as normal data or abnormal data.

[0033] As a further solution of the present invention: the data analysis module includes:

[0034] Heating loss assessment unit: Dynamically assess the heating loss coefficient through the flow rate and temperature data collected by the sampling nodes in the heat source monitoring area, heat transfer monitoring area and heat dissipation monitoring area;

[0035] Heating demand analysis unit: Based on the pressure, flow rate and temperature data collected by the sampling nodes in the heat dissipation monitoring area, as well as the historical heating records, analyze the heating demand of users and predict the heating quantity required in the heat dissipation monitoring area;

[0036] Heating regulation unit: Adjust the heating parameters according to the real-time prediction result of the heating demand and the heating loss coefficient.

[0037] As a further solution of the present invention: Dynamically assess the heating loss coefficient through the flow rate and temperature data collected by the sampling nodes in the heat source monitoring area, heat transfer monitoring area and heat dissipation monitoring area, including the following steps:

[0038] Obtain the subset data of the heat transfer monitoring area and heat dissipation monitoring area corresponding to the heat source monitoring area;

[0039] According to the flow rate data of each pipeline sampling node in the heat transfer monitoring area and heat dissipation monitoring area, calculate the mean value of all pipeline flow rate detection data and use it as the pipeline flow rate data. Dynamically assess the heating loss coefficient through the following formula:

[0040]

[0041]

[0042]

[0043] Among them, Rloss is the heat supply loss coefficient, and Rs is the heat supply loss coefficient from the heat source monitoring area to the heat transfer monitoring area; L0 is the total flow rate at the outlet of the heat source monitoring area, L1 is the sum of the flow rate data of each pipeline in the heat transfer monitoring area, T0 is the temperature at the outlet of the heat source monitoring area, and T1 is the average value of the temperature data of each pipeline in the heat transfer monitoring area; Rz is the heat supply loss coefficient of the heat dissipation monitoring area in the heat transfer monitoring area, L2 is the total flow rate at the inlet of the heat dissipation monitoring area, T2 is the temperature at the inlet of the heat dissipation monitoring area, and w1 and w2 are the weights of different heat supply loss coefficients respectively.

[0044] As a further solution of the present invention: Based on the pressure, flow rate, and temperature data collected by the sampling nodes in the heat dissipation monitoring area, as well as historical heating records, analyze the heating demand of users and predict the required heat supply in the heat dissipation monitoring area, including the following steps:

[0045] Obtain the pressure, flow rate, and temperature data collected by the sampling nodes in the heat dissipation monitoring area at the same time period in different years in the historical heat supply data, as well as the heat supply amount and heating time in the historical heating records;

[0046] Divide the historical data into different time intervals, and successively use the pressure, flow rate, and temperature data of the time interval as inputs, and the heat supply amount data of the corresponding subsequent time interval as outputs to train the RNN recurrent neural network model;

[0047] By inputting the pressure, flow rate, and temperature data collected by the sampling nodes in the heat dissipation monitoring area in real time into the trained RNN recurrent neural network model, predict the required heat supply in the heat dissipation monitoring area.

[0048] As a further solution of the present invention: Adjust the heating parameters according to the real-time prediction result of the heating demand and the heat supply loss coefficient, including the following steps:

[0049] According to the real-time prediction result of the heating demand and the heat supply loss coefficient, and obtain the temperatures at the outlet and return water inlet of the heat source monitoring area, and calculate the target temperature at the outlet of the heat source monitoring area through the following formula:

[0050]

[0051] Among them, Tout is the target temperature at the outlet of the heat source monitoring area, Qout is the predicted value of the required heat supply in the heat dissipation monitoring area, Rloss is the heat supply loss coefficient, Qde is the current heat supply in the heat dissipation monitoring area, m is the flow rate at the heat source outlet of the heat source monitoring area, unit: kg / s, c is the specific heat capacity of water, and Tin is the temperature at the heat source return water inlet of the heat source monitoring area;

[0052] Adjust the water temperature of the outlet of the heating parameters according to the target temperature at the outlet of the heat source monitoring area.

[0053] Compared with the prior art, the beneficial effects of the present invention are:

[0054] In the present invention, the data segmentation processing module divides the heating data into different subsets according to the start-up stage, stable operation stage, and shutdown stage of the heating system; by extracting features from each subset of the historical heating data, a normal data recognition model for each stage is established; there may be significant differences in the heating characteristics and demands in the start-up stage, stable operation stage, and shutdown stage. By modeling each stage separately, these differences can be better captured, thereby improving the prediction accuracy of the model; at the same time, in the stable operation stage, the state of the system is relatively stable and the data noise is less, which helps to improve the ability of the model to identify the normal state.

[0055] In the present invention, the data analysis module dynamically evaluates the heating loss coefficient through the data collected by the sampling nodes in the heat source monitoring area, the heat transfer monitoring area, and the heat dissipation monitoring area. At the same time, it makes a real-time prediction of the user's heating demand, and adjusts the heating parameters according to the real-time prediction result of the heating demand and the heating loss coefficient. According to the real-time evaluated heating loss coefficient and user demand, the heating parameters can be accurately adjusted, thereby reducing energy waste and improving the overall energy utilization efficiency. While meeting the user's demand, overheating is avoided and the overall energy consumption is reduced. Through real-time monitoring and dynamic adjustment, system failures caused by overload or insufficiency can be prevented, thereby improving the reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0057] Figure 1 It is a schematic diagram of the system framework of the present invention;

[0058] Figure 2 It is a schematic diagram of the framework of the data analysis module of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0059] The following will clearly and completely describe the technical solutions of the present invention in combination with the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0060] Please refer to Figure 1 - Figure 2 , an embodiment of the first aspect of the present invention provides a centralized heating data analysis system for decentralized users based on the Internet of Things, including:

[0061] Intelligent zoning module: Divide the heating system into a heat source monitoring area, a heat transfer monitoring area, and a heat dissipation monitoring area. Set sampling nodes at the heat output positions in the heat source monitoring area. Divide the heat transfer monitoring area into an underground transfer area and a floor transfer area, and set sampling nodes with different densities in the underground transfer area and the floor transfer area. Set terminal data sampling nodes in the heat dissipation monitoring area;

[0062] Data acquisition module: Collect heating data at the sampling nodes in each area of the heating system; among them, collect pipeline temperature, pressure, sound, and flow data in the underground transfer area of the heat transfer monitoring area, and collect pipeline temperature, pressure, sound, video, and flow data in the floor transfer area;

[0063] Data segmentation processing module: Divide the heating data into different subsets according to the start-up stage, stable operation stage, and shutdown stage of the heating system; establish a normal data recognition model for each stage by extracting features from each subset of historical heating data;

[0064] Real-time data comparison module: Identify the real-time data with the normal data recognition model of the corresponding stage, and perform abnormal recognition analysis on the abnormal data;

[0065] Data analysis module: Dynamically evaluate the heating loss coefficient through the data collected by the sampling nodes in the heat source monitoring area, the heat transfer monitoring area, and the heat dissipation monitoring area. At the same time, make a real-time prediction of the heating demand of users, and adjust the heating parameters according to the real-time prediction result of the heating demand and the heating loss coefficient.

[0066] Specifically, in this embodiment, the heat source monitoring area: mainly monitors the output of the heat source equipment, and sets sampling nodes to collect temperature, pressure, and flow data in real time.

[0067] Heat transfer monitoring area: Divided into an underground transfer area and a floor transfer area. Set sampling nodes with different densities in these two areas to more accurately monitor the losses and efficiencies in the heat transfer process.

[0068] Heat dissipation monitoring area: Mainly focuses on the performance of radiators or terminal devices. Set terminal data sampling nodes in this area to obtain temperature and flow data in real time.

[0069] By using wireless communication technologies such as LoRa, Zigbee, Wi-Fi, etc., the collected data is sent to the central control system or cloud server in real time for unified processing and analysis.

[0070] The heating data is divided into different subsets by the data segmentation processing module according to the start-up stage, stable operation stage, and shutdown stage of the heating system; by extracting features from each subset of historical heating data, a normal data recognition model for each stage is established; there may be significant differences in the heating characteristics and demands in the start-up stage, stable operation stage, and shutdown stage. By modeling each stage separately, these differences can be better captured, thereby improving the prediction accuracy of the model; at the same time, in the stable operation stage, the state of the system is relatively stable and the data noise is less, which helps to improve the model's ability to recognize the normal state.

[0071] By identifying the normal data in different stages, it is convenient to quickly adjust the heating strategy dynamically according to the current system state, improve energy utilization efficiency, and reduce operating costs. Establishing a normal data recognition model facilitates the timely discovery of abnormal situations, prevents equipment failures or excessive energy consumption, and thus realizes proactive maintenance. Timely discovering and handling potential problems can reduce system downtime and improve the availability of the overall heating system.

[0072] The real-time data comparison module identifies the real-time data with the normal data recognition model of the corresponding stage, and performs abnormal recognition analysis on the abnormal data; through the comparative analysis of real-time monitoring and the normal data model, abnormal data deviating from the normal range can be quickly identified, thus early discovering potential failures or abnormal operation of the system.

[0073] Subsequently, based on the data identified as abnormal, it is convenient to implement a predictive maintenance strategy, perform necessary inspections and repairs before a failure occurs, and thus reduce downtime.

[0074] The data analysis module dynamically evaluates the heating loss coefficient through the data collected by the sampling nodes in the heat source monitoring area, heat transfer monitoring area, and heat dissipation monitoring area. At the same time, it real-time predicts the heating demand of users, and adjusts the heating parameters according to the real-time prediction results of the heating demand and the heating loss coefficient.

[0075] According to the real-time evaluated heating loss coefficient and user demand, the heating parameters can be accurately adjusted, thereby reducing energy waste and improving the overall energy utilization efficiency. While meeting the user demand, overheating is avoided and the overall energy consumption is reduced. Through real-time monitoring and dynamic adjustment, system failures caused by overload or insufficiency can be prevented, thus improving the reliability of the system.

[0076] In one embodiment of the present invention, the heat transfer monitoring area is divided into an underground transfer area and a floor transfer area, and sampling nodes with different densities are set in the underground transfer area and the floor transfer area, including the following steps:

[0077] Set the transmission pipeline with the heat transfer monitoring area buried underground as the underground transmission area, and set the transmission pipeline between the above-ground buildings as the floor transmission area;

[0078] For the underground transmission area, set a sampling node every 500 meters to collect pipeline temperature, pressure, sound and flow data;

[0079] For the floor transmission area, set a sampling node every 10 meters to collect pipeline temperature, pressure, sound and flow data, and set video sampling nodes at the positions of the water separator, water collector and floor main valve.

[0080] In one embodiment of the present invention, heat supply data is collected at the sampling nodes in each area of the heat supply system, including the following steps:

[0081] Collect the pressure, flow and temperature data of the heat source outlet and return water through the sampling nodes in the heat source monitoring area;

[0082] Collect pipeline temperature, pressure, sound, video and flow data through the sampling nodes in the heat transfer monitoring area;

[0083] Collect the pressure, flow and temperature data of the heat dissipation terminal inlet and return water through the sampling nodes in the heat dissipation monitoring area;

[0084] Number the pipelines and sampling nodes in each area, and add data labels to the heat supply data collected by the sampling nodes according to the sampling pipeline and node numbers, and the sampling timestamp.

[0085] Specifically, in this embodiment, establish a numbering rule: formulate a unified numbering rule for the pipeline points in each area. The numbering should be concise and clear for easy identification. For example: Heat source monitoring area: H1, H2, H3...; Heat transfer monitoring area: T1, T2, T3...; Heat dissipation monitoring area: R1, R2, R3...; And set numbers for the sampling points, and add data labels to the heat supply data collected by the sampling nodes according to the sampling pipeline and node numbers, and the sampling timestamp. For example, the data label added to the heat supply data in the heat transfer monitoring area is: T1-001,2023-10-01T12:00:00Z.

[0086] In one embodiment of the present invention, the data segmentation processing module divides the heat supply data into different subsets according to the start-up stage, stable operation stage and shutdown stage of the heat supply system, including the following steps:

[0087] Set the stage from the shutdown state of the heat supply system to the stage when the temperature detected by the sampling node in the heat dissipation monitoring area reaches the predetermined temperature as the start-up stage;

[0088] After the temperature detected by the sampling node in the heat dissipation monitoring area of the heating system reaches the predetermined temperature and before the heat source in the heat source monitoring area stops heating, this stage is set as the stable operation stage;

[0089] The process from the heat source in the heat source monitoring area of the heating system stopping heating to the flow rate in the pipeline in the heat dissipation monitoring area dropping to the preset threshold is set as the shutdown stage;

[0090] According to the settings of the startup stage, stable operation stage and shutdown stage of the heating system, and based on the timestamps of the heating data, the heating data is divided into different subsets, and according to the data tags in each subset, the data in the subset is partitioned, and the data belonging to the same sampling pipeline in the subset is assigned to the same partition.

[0091] In one embodiment of the present invention, by extracting features from each subset of historical heating data, a normal data recognition model for each stage is established, including the following steps:

[0092] Obtain historical heating data, label the data when the heating system operates normally in different areas as normal data, label the data when the heating system operates abnormally in different areas as abnormal data, and divide the historical heating data into different data subsets according to the acquisition stage of the historical heating data;

[0093] Extract the sound and video data of the pipeline collected by the sampling node in the heat transfer monitoring area from the historical heating data subset;

[0094] Extract the time-domain features of the audio of the pipeline sound data, including: mean, variance and peak value, convert the time-domain signal of the audio into a frequency-domain representation through Fourier transform, and extract the frequency-domain features, including: frequency width and power spectral density;

[0095] Mark the pictures with abnormalities in the image data of the positions of the water distributor, water collector and floor main valve in the historical data, and train the deep learning model with the marked image data to identify whether the images of the positions of the water distributor, water collector and floor main valve are abnormal;

[0096] For the video data of the positions of the water distributor, water collector and floor main valve collected by the sampling node in the heat transfer monitoring area, extract one image frame every 50 frames, and input the extracted image frames into the trained deep learning model to identify whether abnormalities occur;

[0097] Replace the sound data of the pipeline collected by the sampling node in the heat transfer monitoring area with the time-domain features and the sound data of the pipeline in the frequency-domain feature replacement subset, and replace the video data in the heat transfer monitoring area with the judgment result of whether the image is abnormal;

[0098] Train an MLP (Multi-Layer Perceptron) using subset data from the start-up phase, stable operation phase, and shutdown phase of the heating system respectively, and identify the subset data from the start-up phase, stable operation phase, and shutdown phase as normal data or abnormal data.

[0099] In one embodiment of the present invention, the data analysis module includes:

[0100] Heating loss assessment unit: Dynamically evaluate the heating loss coefficient through the flow rate and temperature data collected by the sampling nodes in the heat source monitoring area, heat transfer monitoring area, and heat dissipation monitoring area.

[0101] Heating demand analysis unit: Analyze the heating demand of users and predict the heat supply required in the heat dissipation monitoring area based on the pressure, flow rate, and temperature data collected by the sampling nodes in the heat dissipation monitoring area, as well as the historical heating records.

[0102] Heating adjustment unit: Adjust the heating parameters according to the real-time prediction result of the heating demand and the heating loss coefficient.

[0103] In one embodiment of the present invention, dynamically evaluating the heating loss coefficient through the flow rate and temperature data collected by the sampling nodes in the heat source monitoring area, heat transfer monitoring area, and heat dissipation monitoring area includes the following steps:

[0104] Obtain the subset data of the heat transfer monitoring area and heat dissipation monitoring area corresponding to the heat source monitoring area;

[0105] According to the flow rate data of each pipeline sampling node in the heat transfer monitoring area and heat dissipation monitoring area, calculate the average value of all pipeline flow rate detection data and use it as the pipeline flow rate data. Dynamically evaluate the heating loss coefficient through the following formula:

[0106]

[0107]

[0108]

[0109] Where, Rloss is the heating loss coefficient, Rs is the heating loss coefficient from the heat source monitoring area to the heat transfer monitoring area; L0 is the total flow rate at the outlet of the heat source monitoring area, L1 is the sum of the flow rate data of each pipeline in the heat transfer monitoring area, T0 is the temperature at the outlet of the heat source monitoring area, T1 is the average value of the temperature data of each pipeline in the heat transfer monitoring area; Rz is the heating loss coefficient from the heat transfer monitoring area to the heat dissipation monitoring area, L2 is the total flow rate at the inlet of the heat dissipation monitoring area, T2 is the temperature at the inlet of the heat dissipation monitoring area, and w1 and w2 are the weights of different heating loss coefficients respectively.

[0110] Specifically, in this embodiment, by obtaining the data statistics of the total heat output in the heat source monitoring area in historical data, and the data statistics of the total heat obtained from the heat dissipation monitoring area corresponding to the heat source in the heat source monitoring area, calculating the actual heat loss value, and comparing the ratio of the actual heat loss value to the total heat output in the heat source monitoring area with the heat supply loss coefficient calculated by the above formula, when w1 is set to 0.67 and w2 is set to 0.33.

[0111] In one embodiment of the present invention, based on the pressure, flow rate, and temperature data collected by the sampling nodes in the heat dissipation monitoring area, and historical heating records, analyze the heating demand of users and predict the heat supply required for the heat dissipation monitoring area, including the following steps:

[0112] Obtain the pressure, flow rate, and temperature data collected by the sampling nodes in the heat dissipation monitoring area at the same time in different years in the historical heating data, as well as the heat supply and heating time in the historical heating records;

[0113] Divide the historical data into different time intervals, and successively use the pressure, flow rate, and temperature data of the time interval as inputs, and the heat supply data of the corresponding subsequent time interval as outputs to train the RNN recurrent neural network model;

[0114] By inputting the pressure, flow rate, and temperature data collected by the sampling nodes in the heat dissipation monitoring area collected in real time into the trained RNN recurrent neural network model, predict the heat supply required for the heat dissipation monitoring area.

[0115] In one embodiment of the present invention, according to the real-time prediction result of the heating demand and the heat supply loss coefficient, adjust the heating parameters, including the following steps:

[0116] According to the real-time prediction result of the heating demand and the heat supply loss coefficient, and obtain the temperatures at the water outlet and water return of the heat source monitoring area, and calculate the target temperature at the water outlet of the heat source monitoring area through the following formula:

[0117]

[0118] Among them, Tout is the target temperature at the water outlet of the heat source monitoring area, Qout is the predicted value of the heat supply required for the heat dissipation monitoring area, Rloss is the heat supply loss coefficient, Qde is the current heat supply of the heat dissipation monitoring area, m is the flow rate at the water outlet of the heat source in the heat source monitoring area, unit: kg / s, c is the specific heat capacity of water, and Tin is the temperature at the water return of the heat source in the heat source monitoring area;

[0119] According to the target temperature at the water outlet of the heat source monitoring area, adjust the water temperature at the water outlet of the heating parameters.

[0120] The above embodiments are only used to illustrate the technical method of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A centralized heating data analysis system for decentralized users based on the Internet of Things, characterized in that: include: Intelligent zoning module: the heating system is divided into heat source monitoring area, heat transmission monitoring area and heat dissipation monitoring area, sampling nodes are set at the heat output position of the heat source monitoring area, the heat transmission monitoring area is divided into underground transmission area and floor transmission area, sampling nodes of different densities are set in the underground transmission area and floor transmission area, and terminal data sampling nodes are set in the heat dissipation monitoring area; Data collection module: collects heating data at sampling nodes in various areas of the heating system; collects pipeline temperature, pressure, sound and flow data in the underground transmission area of ​​the heat transmission monitoring area, and collects pipeline temperature, pressure, sound, video and flow data in the floor transmission area; Data segmentation processing module: The heating data is divided into different subsets according to the startup phase, stable operation phase and shutdown phase of the heating system; by extracting features from each subset of the historical heating data, a normal data recognition model for each phase is established; the sound data of the pipelines in the subset is extracted for time domain features and frequency domain features, and the original sound data is replaced; by training a deep learning model, the video data of the water distributor, water collector and floor master valve positions collected by the sampling nodes in the thermal transmission monitoring area are detected to determine whether an abnormality occurs, and the judgment result replaces the original video data; the MLP multilayer perceptron is trained through the subset data of the startup phase, stable operation phase and shutdown phase of the heating system to identify the subset data of the startup phase, stable operation phase and shutdown phase as normal data or abnormal data; Real-time data comparison module: identifies real-time data with the normal data recognition model of the corresponding stage, and performs abnormal recognition analysis on abnormal data; Data analysis module: dynamically evaluates the heat loss coefficient through the data collected by the sampling nodes in the heat source monitoring area, heat transmission monitoring area and heat dissipation monitoring area. At the same time, it makes real-time predictions of the user's heating demand and adjusts the heating parameters according to the real-time prediction results of the heating demand and the heat loss coefficient. The data segmentation processing module divides the heating data into different subsets according to the startup phase, stable operation phase and shutdown phase of the heating system, including the following steps: The stage from the heating system being in the off state to the stage when the sampling node in the heat dissipation monitoring area detects that the temperature reaches the predetermined temperature is set as the startup stage; The stage from when the temperature of the heating system is detected by the sampling node in the heat dissipation monitoring area to when the temperature reaches the predetermined temperature to when the heat source in the heat source monitoring area stops supplying heat is set as the stable operation stage; The process from when the heat source in the heat source monitoring area stops supplying heat to when the flow rate in the pipeline in the heat dissipation monitoring area drops to a preset threshold value is set as the shutdown stage; According to the settings of the startup phase, stable operation phase and shutdown phase of the heating system, the heating data is divided into different subsets according to the timestamp of the heating data, and the data in the subset is partitioned according to the data label in each subset, and the data belonging to the same sampling pipeline in the subset is divided into the same partition.

2. According to claim 1, a centralized heating data analysis system for dispersed users based on the Internet of Things is characterized in that: The thermal transmission monitoring area is divided into an underground transmission area and a floor transmission area, and sampling nodes of different densities are set in the underground transmission area and the floor transmission area, including the following steps: The transmission pipelines buried underground in the heat transmission monitoring area are set as underground transmission areas, and the transmission pipelines located between above-ground buildings are set as floor transmission areas; For underground transmission areas, a sampling node is set up every 500 meters to collect pipeline temperature, pressure, sound and flow data; For the floor transmission area, a sampling node is set up every 10 meters to collect pipeline temperature, pressure, sound and flow data, and video sampling nodes are set up at the water distributor, water collector and floor main valve positions.

3. According to the Internet of Things-based decentralized heating data analysis system of claim 1, it is characterized in that: Heating data collection is performed at sampling nodes in various areas of the heating system, including the following steps: The pressure, flow rate and temperature data of heat source outlet and return water are collected through sampling nodes in the heat source monitoring area; Collect pipeline temperature, pressure, sound, video and flow data through sampling nodes in the thermal transmission monitoring area; The pressure, flow and temperature data of the inlet and return water of the heat dissipation terminal are collected through the sampling nodes in the heat dissipation monitoring area; The pipelines and sampling nodes in each area are numbered, and data labels are added to the heating data collected by the sampling nodes according to the sampling pipeline and node numbers and the sampling timestamp.

4. According to the Internet of Things-based decentralized heating data analysis system of claim 1, it is characterized in that: By extracting features from each subset of historical heating data, a normal data recognition model for each stage is established, including the following steps: Obtain historical heating data, and mark the data of different areas of the heating system when they are operating normally as normal data, and mark the data of different areas of the heating system when they are operating abnormally as abnormal data, and divide the historical heating data into different data subsets according to the collection stage of the historical heating data; From the historical heating data subset, extract the sound and video data of the sampling nodes and collection pipelines in the heat transmission monitoring area; Extract the time domain features of the audio data of the pipeline, including mean, variance and peak value, convert the time domain signal of the audio into frequency domain representation through Fourier transform, and extract frequency domain features, including frequency width and power spectrum density; Mark the abnormal images in the image data of the water distributor, water collector and floor master valve position in the historical data, train the deep learning model with the marked image data, and identify whether the images of the water distributor, water collector and floor master valve position are abnormal; For the video data of the water distributor, water collector and floor master valve positions collected by the sampling nodes in the heat transfer monitoring area, an image frame is extracted every 50 frames, and the extracted image frame is input into the trained deep learning model to identify whether there is an abnormality; The sound data of the pipeline collected by the sampling node in the heat transfer monitoring area is replaced with the time domain features and the frequency domain features, the sound data of the pipeline in the subset is replaced, and the video data of the heat transfer monitoring area is replaced with the judgment result of whether the image is abnormal; The MLP multilayer perceptron is trained by using subset data of the startup phase, stable operation phase and shutdown phase of the heating system to identify the subset data of the startup phase, stable operation phase and shutdown phase as normal data or abnormal data.

5. According to the Internet of Things-based decentralized user central heating data analysis system of claim 1, it is characterized in that: The data analysis module comprises: Heat loss assessment unit: dynamically assesses the heat loss coefficient through the flow and temperature data collected by sampling nodes in the heat source monitoring area, heat transfer monitoring area, and heat dissipation monitoring area; Heating demand analysis unit: Based on the pressure, flow and temperature data collected by the sampling nodes in the heat dissipation monitoring area, as well as the historical heating records, it analyzes the user's heating demand and predicts the required heating amount in the heat dissipation monitoring area; Heating regulation unit: adjusts heating parameters according to the real-time forecast results of heating demand and the heat loss coefficient.

6. The centralized heating data analysis system for dispersed users based on the Internet of Things according to claim 5 is characterized in that: The heat loss coefficient is dynamically evaluated through the flow and temperature data collected by the sampling nodes in the heat source monitoring area, the heat transfer monitoring area, and the heat dissipation monitoring area, including the following steps: Obtain subset data of the heat transfer monitoring area and the heat dissipation monitoring area corresponding to the heat source monitoring area; According to the flow data of each pipeline sampling node in the heat transmission monitoring area and the heat dissipation monitoring area, the mean of all pipeline flow detection data is calculated and used as pipeline flow data to dynamically evaluate the heat loss coefficient through the following formula: Among them, Rloss is the heat loss coefficient, Rs is the heat loss coefficient from the heat source monitoring area to the thermal transmission monitoring area; L0 is the total flow at the water outlet of the heat source monitoring area, L1 is the sum of the flow data of each pipeline in the thermal transmission monitoring area, T0 is the temperature at the water outlet of the heat source monitoring area, T1 is the mean of the temperature data of each pipeline in the thermal transmission monitoring area; Rz is the heat loss coefficient of the heat dissipation monitoring area in the thermal transmission monitoring area, L2 is the total flow at the water inlet of the heat dissipation monitoring area, T2 is the temperature at the water inlet of the heat dissipation monitoring area, and w1 and w2 are the weights of different heat loss coefficients respectively.

7. The centralized heating data analysis system for dispersed users based on the Internet of Things according to claim 5 is characterized in that: Based on the pressure, flow and temperature data collected by the sampling nodes in the heat dissipation monitoring area and the historical heating records, the user's heating demand is analyzed and the required heating amount in the heat dissipation monitoring area is predicted, including the following steps: Obtain the pressure, flow and temperature data collected by the sampling nodes in the heat dissipation monitoring area in different years and the same period in the historical heating data, as well as the heat supply and heating time in the historical heating records; The historical data is divided into different time intervals, and the pressure, flow and temperature data of the time interval are used as input in turn, and the corresponding heating data of the next time interval is used as output to train the RNN recurrent neural network model; The pressure, flow and temperature data collected by the sampling nodes in the heat dissipation monitoring area in real time are input into the trained RNN recurrent neural network model to predict the required heating supply in the heat dissipation monitoring area.

8. The centralized heating data analysis system for dispersed users based on the Internet of Things according to claim 5 is characterized in that: According to the real-time forecast results of the heating demand and the heat loss coefficient, the heating parameters are adjusted, including the following steps: According to the real-time forecast results of heating demand and the heat loss coefficient, and the temperatures at the water outlet and return water outlet of the heat source monitoring area are obtained, the target temperature at the water outlet of the heat source monitoring area is calculated by the following formula: Among them, Tout is the target temperature at the water outlet of the heat source monitoring area, Qout is the predicted value of the heat supply required in the heat dissipation monitoring area, Rloss is the heat loss coefficient, Qde is the current heat supply in the heat dissipation monitoring area, m is the flow rate at the water outlet of the heat source in the heat source monitoring area, unit: kg / s, c is the specific heat capacity of water, Tin is the temperature at the return water outlet of the heat source in the heat source monitoring area; According to the target temperature at the water outlet of the heat source monitoring area, the outlet water temperature of the heating parameter is adjusted.

Images