A heating data processing system and method for a heating system

Abstract

The present invention discloses a heating data processing system and method for a heating system, relating to the technical field of heating data processing. It solves the technical problems that it is difficult to classify and summarize the data in each node of the heating system and then perform classification processing, and it is also difficult to establish connections between data of nodes with high similarity. By using wireless communication devices to establish a data transmission channel between edge data nodes, it is convenient to achieve low-latency data transmission, enabling the system to respond to changes in real time. Establishing a data transmission channel between adjacent edge nodes can effectively disperse the network traffic load and improve the stability of the overall network. Establishing a data table at the edge data node at the heat source helps to centrally manage data from different terminals and nodes, enhancing data accessibility and management efficiency. By correlating data with high similarity, it is convenient to analyze the relationship between highly similar nodes in subsequent data mining, providing deeper information for subsequent analysis.

Description

Technical Field

[0001] The present invention belongs to the field of heating data processing, and specifically relates to a heating data processing system and method for a heating system. Background Art

[0002] With the acceleration of the urbanization process and the improvement of people's living standards, the heating demand is constantly increasing. Especially in cold regions, the efficiency and reliability of heating systems are particularly important. The global increasing attention to reducing carbon emissions and improving energy utilization efficiency has promoted the transformation and optimization of traditional heating systems. Through data analysis, energy use can be better monitored and managed, thereby reducing the environmental impact. The development of information technology enables emerging technologies such as the Internet of Things (IoT), big data, and artificial intelligence (AI) to be applied to heating systems to achieve intelligent monitoring and management and improve the automation level of the systems. The heat data processing system of a heating system is an important tool for realizing modern intelligent heating management. Through effective data processing, it is convenient for subsequent data analysis.

[0003] In most heat data processing solutions for heating systems, it is difficult to classify and summarize the data in each node of the heating system and then perform classification processing, and it is also difficult to establish connections between node data with high similarity by overall statistical analysis of the data of each adopted node. 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 proposes a heating data processing system and method for a heating system, which are used to solve the technical problems that it is difficult to classify and summarize the data in each node of the heating system and then perform classification processing, and it is also difficult to establish connections between node data with high similarity.

[0005] To solve the above problems, the first aspect of the present invention provides a heating data processing system for a heating system, including:

[0006] Node setting module: Edge data nodes are set at the transmission pipeline network and heat source in the heating system. According to the grouping principle of grouping the terminals in the heating system into the same terminal group based on the terminals connected to the same branch of the transmission pipeline network, the terminals in the heating system are grouped, and data aggregation nodes are set on the same branch of the transmission pipeline network connected to the terminals in each group;

[0007] Transmission channel construction module: A data mirror warehouse and a wireless communication device are set at each edge data node. The edge data nodes are numbered, and data tags are set for the data according to the time stamp of data collection and the node number. A data transmission channel is established in sequence between adjacent edge data nodes from the edge data node at the heat source to the data aggregation node;

[0008] Terminal Data Aggregation Module: The terminal data collected by each terminal in the terminal group is sent to the data aggregation node in the group. After the data aggregation node stores the data, it is packaged and sent to the nearest edge data node;

[0009] Transportation Management Module: After the edge data node collects the data of this node, it stores the data of this node in the mirror repository, and packages the data of this node and the terminal data packet together and transmits them through the data transmission channel between the edge data nodes to the edge data node at the heat source;

[0010] Data Association Module: The edge data node at the heat source standardizes the received data, establishes a data table for the node, and associates the data with high similarity.

[0011] As a further solution of the present invention: Set up a data mirror repository and a wireless communication device at each edge data node, and number the edge data nodes, including the following steps:

[0012] Set up a wireless communication device and a data mirror repository at each edge data node. Among them, the data mirror repository is used to store the data copy collected by the edge node; Select a suitable data format, such as JSON, CSV or binary format, for convenient subsequent processing and analysis.

[0013] The edge data node is set with a wireless communication device, and a Wi-Fi module suitable for short-distance and high-bandwidth data transmission, or a LoRa module suitable for long-distance and low-power data transmission, or a Zigbee module suitable for low-power and short-distance network applications can be used. Specifically, it can be selected according to the network passing environment at the node setting location.

[0014] Assign a unique identifier to each edge data node as the edge data node number, create a transmission pipeline node management diagram, and record the location of each edge node in the transmission pipeline, the node number, and the number of nodes with the least interval from the edge data node at the heat source.

[0015] As a further solution of the present invention: From the edge data node at the heat source to the data aggregation node direction, sequentially establish a data transmission channel between adjacent edge data nodes, including the following steps:

[0016] From the edge data node at the heat source to the data aggregation node direction, according to the transmission pipeline node management diagram, sequentially establish a data transmission channel between adjacent edge data nodes through the wireless communication device on the transmission pipeline;

[0017] The aggregation node establishes a data transmission channel with the nearest edge data node through the MQTT communication protocol.

[0018] As a further solution of the present invention: After the edge data nodes at the heat source standardize the received data, they establish a data table for the nodes and associate the data with high similarity, including the following steps:

[0019] The edge data nodes at the heat source summarize the data sent by the edge data nodes and send the data together with the data of this node to the cloud management platform. The cloud management platform standardizes the data of each edge data node and establishes a data table for the terminal data and a data table for the transmission pipeline network nodes;

[0020] According to the data collected by the terminals of the heating system and the node data collected by the edge data nodes on the transmission pipeline network, a similarity evaluation model between terminal nodes and a similarity evaluation model for transmission pipeline network nodes are constructed;

[0021] The similarity between terminal nodes evaluated by the similarity evaluation model between terminal nodes and the similarity between transmission pipeline network nodes evaluated by the similarity evaluation model for transmission pipeline network nodes;

[0022] Associate the data between terminals with high similarity and associate the data between transmission pipeline network nodes with high similarity.

[0023] As a further solution of the present invention: The cloud management platform standardizes the data of each edge data node and establishes a data table for the terminal data and a data table for the transmission pipeline network nodes, including the following steps:

[0024] Establish a data standardization rule for the edge data nodes and convert the data collected by each edge data node into the same format;

[0025] According to the node data collected by the edge data nodes, including: flow data, heat data, and sound data, where the heat data includes: internal pipe temperature data and external pipe temperature data, and the sound data includes external pipe sound audio data and internal pipe sound audio data;

[0026] The edge data nodes at the heat source collect node data as the flow data, heat data, and sound data at the water outlet and return water of the heat source;

[0027] Establish a data table for the transmission pipeline network nodes, including: a data list for monitoring flow data, heat data, and sound data, and establish a connection of the same node data between each data list according to the data tags of the node data in the data list;

[0028] Set terminal nodes to collect data at the terminals of the heating system, number the terminal nodes, and add data tags to the data of the terminal nodes according to the terminal node numbers and terminal groups;

[0029] According to the data collected by the terminal node, including: flow data, heat data, pressure data, and geographical information data, where the heat data includes: the ambient temperature of the terminal and the temperature inside the pipeline, and the geographical information data includes: the geographical coordinates and floor location of the terminal node;

[0030] Establish a terminal data table, including: data lists of flow data, heat data, pressure data, and geographical information data, and establish connections for the same terminal group data between each data list according to the data tags of the node data in the data list.

[0031] As a further solution of the present invention: construct a similarity evaluation model between terminal nodes, including the following steps:

[0032] Obtain the terminal pipeline topology data according to the geographical information data of the terminal, the flow data, heat data, pressure data, and geographical information data collected by the terminal node; where the terminal pipeline topology data includes: the connecting pipelines and locations of the terminal node in the transmission pipeline network node management diagram, and the number of nodes directly connected to the terminal node;

[0033] Establish a terminal pipeline topology data table and add it to the established terminal data table;

[0034] Construct a comprehensive similarity recognition model for flow data and heat data through a deep learning algorithm;

[0035] Evaluate the comprehensive similarity of flow data and heat data between terminal nodes through the comprehensive similarity recognition model of flow data and heat data;

[0036] According to the comprehensive similarity evaluation result, as well as the terminal pipeline topology data table, and the data tables of pressure data and geographical information data, establish a similarity evaluation model between terminal nodes through the following formula:

[0037]

[0038] Where S is the similarity evaluation value between terminal nodes, S1 is the similarity of the terminal pipeline topology data between terminal nodes, S2 is the similarity of the pressure data between terminal nodes, S3 is the similarity of the geographical information data between terminal nodes, C is a constant set according to the comprehensive similarity evaluation result, w1, w2, and w3 are the weights of the similarity of the terminal pipeline topology data between terminal nodes, the similarity of the pressure data between terminal nodes, and the similarity of the geographical information data between terminal nodes respectively; g0 is the average value of the number of nodes directly connected to the terminal nodes in the terminal group where the two terminal nodes are located, is the difference in the number of nodes directly connected to two terminal nodes, Dg0 is the average distance between the terminal nodes and the connecting pipelines in the transmission pipeline network node management diagram in the terminal group where the two terminal nodes are located, Dg1 is the distance between the terminal node and the connecting pipelines in the transmission pipeline network node management diagram, and Dg2 is the distance between the other terminal node and the connecting pipelines in the transmission pipeline network node management diagram; S0 is the average variance of the terminal node pressure data in the preset time period in the terminal group where the two terminal nodes are located, is the difference in variance of the terminal node pressure data in the preset time period, is the average value of the terminal node pressure data in the terminal group where the two terminal nodes are located, is the difference in average value of the terminal node pressure data in the preset time period; is the average distance between the geographical coordinates of the terminal nodes in the terminal group where the two terminal nodes are located, is the distance between the geographical coordinates of the terminal nodes, is the sum of the total number of floors of the two terminal nodes' locations, is the difference in the number of floors where the terminal nodes are located.

[0039] As a further solution of the present invention: construct a transmission pipeline network node similarity evaluation model, including the following steps:

[0040] Obtain the flow data, heat data, and sound data collected by the edge data nodes on the transmission pipeline network;

[0041] Evaluate the comprehensive similarity of the flow data and heat data between the terminal nodes through the comprehensive similarity recognition model of the flow data and heat data;

[0042] Extract the time-domain features and frequency-domain features from the audio data of the transmission pipeline network nodes with high similarity in the transmission pipeline network from the historical data. Through the time-domain features and frequency-domain features, and through machine learning algorithms, train the sound data recognition model to identify whether the transmission pipeline network nodes are high-similarity nodes;

[0043] Among them, extracting the time-domain features of the audio of the sound data of the pipeline includes: 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;

[0044] Identify the similarity of the transmission pipeline network node data through the comprehensive similarity recognition model of the flow data and heat data and the sound data recognition model: if the output data of the comprehensive similarity recognition model of the flow data and heat data, and the output data of the sound data recognition model both identify the node data as high-similarity node data, then the transmission pipeline network node is a high-similarity node; otherwise, it is a non-high-similarity node.

[0045] As a further solution of the present invention: By means of a deep learning algorithm, a comprehensive similarity recognition model for flow data and heat data is constructed, including the following steps:

[0046] From the historical data, for the flow data and heat data of high-similarity nodes in the transmission pipeline network or terminal nodes, the data is sorted into a structured training data set and a validation data set;

[0047] Using the training data set and the validation data set, a recurrent neural network is trained to establish a comprehensive similarity recognition model for flow data and heat data to identify whether the transmission pipeline network nodes are high-similarity nodes.

[0048] As a further solution of the present invention: The similarity between terminal nodes evaluated by a similarity evaluation model between terminal nodes includes the following steps:

[0049] Using the similarity evaluation model between terminal nodes, calculate the similarity evaluation value between terminal nodes;

[0050] If the similarity evaluation value between terminal nodes is greater than a preset threshold, the terminal nodes are high-similarity terminals; otherwise, they are not high-similarity terminals.

[0051] As another solution of the present invention: A method for processing heating data of a heating system includes the following steps:

[0052] Edge data nodes are set at the transmission pipeline network and heat sources in the heating system. According to the grouping principle of grouping terminals in the same terminal group that are connected to the same branch of the transmission pipeline network in the heating system, the terminals in the heating system are grouped, and data aggregation nodes are set on the same branch of the transmission pipeline network connected to the terminals in each group;

[0053] A data mirror warehouse and a wireless communication device are set at each edge data node. The edge data nodes are numbered. According to the time stamp of data collection and the node number, data tags are set for the data. A data transmission channel is established successively between adjacent edge data nodes from the edge data node at the heat source to the data aggregation node;

[0054] The terminal data collected by each terminal in the terminal group is sent to the data aggregation node in the group. After the data aggregation node stores the data, it is packaged and sent to the nearest edge data node;

[0055] After collecting the data of its own node, the edge data node stores the data of its own node in the mirror warehouse, and packages the data of its own node and the terminal data packet and transmits them through the data transmission channel between edge data nodes to the edge data node at the heat source;

[0056] After the edge data nodes at the heat source standardize the received data, they establish a data table for the nodes and correlate the data with high similarity.

[0057] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0058] By setting up a mirror repository at each edge data node, the present invention realizes the local storage and management of data, reduces the dependence on the central server, and improves the data access speed. In case of failures or network interruptions, the edge nodes can quickly restore services using the local mirror repository, ensuring the high availability of the system. By establishing a data transmission channel between edge data nodes through wireless communication devices, it is convenient to achieve low-latency data transmission, enabling the system to respond to changes in real time. By adding a timestamp and a node number to each data packet, the integrity and consistency of the data are ensured. This is for subsequent data analysis and tracking. Establishing a data transmission channel between adjacent edge nodes effectively disperses the network traffic load and improves the overall network stability.

[0059] Through standardization processing, the present invention ensures that all transmitted data adopts a consistent format, facilitating subsequent data analysis and processing. Establishing a data table at the edge data nodes at the heat source helps to centrally manage data from different terminals and nodes, improving data accessibility and management efficiency. Standardizing the received data helps to eliminate deviations caused by different sources or formats, improving the consistency and accuracy of the data. By correlating the data with high similarity, it is convenient to analyze the relationship between highly similar nodes in subsequent data mining, providing deeper information for subsequent analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] 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 following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0061] Figure 1 It is a schematic diagram of the system framework of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0062] The following will clearly and completely describe the technical solutions of the present invention in conjunction 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 fall within the protection scope of the present invention.

[0063] Please refer to Figure 1, an embodiment of the first aspect of the present invention provides a heating data processing system and method for a heating system, including:

[0064] Node setting module: Set edge data nodes at the transmission pipeline network and heat sources in the heating system. According to the grouping principle of grouping terminals connected to the same branch of the transmission pipeline network into the same terminal group, group the terminals in the heating system, and set data aggregation nodes on the same branch of the transmission pipeline network connected to the terminals in each group;

[0065] Transmission channel construction module: Set a data mirror warehouse and a wireless communication device at each edge data node, number the edge data nodes, set data tags for the data according to the time stamp of data collection and the node number, and sequentially establish data transmission channels between adjacent edge data nodes from the edge data node at the heat source to the data aggregation node;

[0066] Terminal data summarization module: The terminal data collected by each terminal in the terminal group is sent to the data aggregation node in the group. After the data aggregation node stores the data, it is packaged and sent to the nearest edge data node;

[0067] Transmission management module: After collecting the data of this node, the edge data node stores the data of this node in the mirror warehouse, and packages and transmits the data of this node and the terminal data packet through the data transmission channel between the edge data nodes to the edge data node at the heat source;

[0068] Data association module: The edge data node at the heat source standardizes the received data, establishes a data table for the node, and associates the data with high similarity.

[0069] Specifically, in this embodiment, by setting edge data nodes at the transmission pipeline network and heat sources in the heating system, according to the grouping principle of grouping terminals connected to the same branch of the transmission pipeline network into the same terminal group, group the terminals in the heating system, and set data aggregation nodes on the same branch of the transmission pipeline network connected to the terminals in each group; by setting data aggregation nodes on the same branch of the transmission pipeline network, it is convenient to effectively concentrate the data from multiple terminals, reduce the complexity of data transmission and bandwidth requirements. The data aggregation node realizes the real-time processing of terminal data, which is convenient for centrally monitoring all terminals in the terminal group and improving the reliability of the system. The centrally processed data provides a basis for subsequent data analysis, which is convenient for summarizing the data collected by the aggregation node; by avoiding each node sending data to the central server separately, the network pressure can be reduced.

[0070] Set up a data mirror repository and a wireless communication device at each edge data node, number the edge data nodes, and set data tags for the data according to the timestamp of data collection and the node number. Establish data transmission channels between adjacent edge data nodes in sequence from the edge data node at the heat source to the data aggregation node;

[0071] By setting up a mirror repository at each edge data node, local storage and management of data are realized, reducing the dependence on the central server and improving the data access speed. In case of failure or network interruption, the edge node can quickly restore services using the local mirror repository, ensuring the high availability of the system. Establish data transmission channels between edge data nodes through wireless communication devices, facilitating low-latency data transmission and enabling the system to respond to changes in real time.

[0072] By adding a timestamp and node number to each data packet, the integrity and consistency of the data are ensured. This is for subsequent data analysis and tracking. Establish data transmission channels between adjacent edge nodes, effectively dispersing the network traffic load and improving the overall network stability.

[0073] Pack the data of this node together with the terminal data packet and transmit it through the data transmission channel between edge data nodes to the edge data node at the heat source; the edge data node at the heat source standardizes the received data, establishes a data table for the node, and correlates the data with high similarity.

[0074] Through standardization processing, ensure that all transmitted data adopts a consistent format, facilitating subsequent data analysis and processing. Establishing a data table at the edge data node at the heat source helps to centrally manage data from different terminals and nodes, improving data accessibility and management efficiency. Standardizing the received data helps to eliminate deviations caused by different sources or formats, improving data consistency and accuracy. By correlating data with high similarity, it is convenient to analyze the relationship between highly similar nodes in subsequent data mining and provide deeper information for subsequent analysis.

[0075] In one embodiment of the present invention, setting up a data mirror repository and a wireless communication device at each edge data node and numbering the edge data nodes include the following steps:

[0076] Set up a wireless communication device and a data mirror repository at each edge data node, where the data mirror repository is used to store copies of data collected by the edge node; select a suitable data format, such as JSON, CSV, or binary format, for convenient subsequent processing and analysis.

[0077] The edge data nodes are equipped with wireless communication devices. Wi-Fi modules suitable for short-distance and high-bandwidth data transmission, LoRa modules suitable for long-distance and low-power data transmission, or Zigbee modules suitable for low-power and short-distance network applications can be adopted, and the specific selection can be based on the network environment at the node setting location.

[0078] Assign a unique identifier to each edge data node as the edge data node number, create a transmission pipeline network node management graph, and record the location of each edge node in the transmission pipeline network, the node number, and the number of nodes with the fewest intervals between the edge data node at the heat source.

[0079] Numbering format: A unified numbering format can be adopted, for example:

[0080] Node_001, Node_002, and so on.

[0081] It can be classified according to the geographical location or function of the node, for example: ZoneA_Node_001.

[0082] In one embodiment of the present invention, in the direction from the edge data node at the heat source to the data aggregation node, a data transmission channel between adjacent edge data nodes is established in sequence, including the following steps:

[0083] In the direction from the edge data node at the heat source to the data aggregation node, according to the transmission pipeline network node management graph, a data transmission channel between adjacent edge data nodes is established on the transmission pipeline network through wireless communication devices in sequence;

[0084] The aggregation node establishes a data transmission channel with the nearest edge data node through the MQTT communication protocol.

[0085] In one embodiment of the present invention, after the edge data node at the heat source standardizes the received data, a data table of the node is established, and the high-similarity data is associated, including the following steps:

[0086] The edge data node at the heat source summarizes the data sent by the edge data nodes and sends it to the cloud management platform together with the data of this node. The cloud management platform standardizes the data of each edge data node and establishes a terminal data table and a data table of the transmission pipeline network nodes;

[0087] According to the data collected by the heating system terminal and the node data collected by the edge data nodes on the transmission pipeline network, a similarity evaluation model between terminal nodes and a similarity evaluation model of transmission pipeline network nodes are constructed;

[0088] The similarity between terminal nodes evaluated by the similarity evaluation model between terminal nodes, and the similarity between transmission pipeline network nodes evaluated by the transmission pipeline network node similarity evaluation model;

[0089] Associate the data between high-similarity terminals and associate the data between high-similarity transmission pipeline network nodes.

[0090] In one embodiment of the present invention, the cloud management platform standardizes the data of each edge data node and establishes a terminal data table and a data table of transmission pipeline network nodes, including the following steps:

[0091] Establish a data standardization rule for edge data nodes to convert the data collected by each edge data node into the same format;

[0092] According to the node data collected by the edge data node, including: flow data, heat data, and sound data, wherein the heat data includes: pipeline internal temperature data and pipeline external temperature data, and the sound data includes pipeline external sound audio data and pipeline internal sound audio data;

[0093] The edge data node at the heat source collects node data as the flow data, heat data, and sound data at the water outlet and return water of the heat source;

[0094] Establish a data table for transmission pipeline network nodes, including: a data list for monitoring flow data, heat data, and sound data, and establish connections of the same node data between each data list according to the data labels of the node data in the data list;

[0095] Set terminal nodes at the heat supply system terminal to collect data, number the terminal nodes, and add data labels to the data of the terminal nodes according to the terminal node numbers and terminal groups;

[0096] According to the data collected by the terminal nodes, including: flow data, heat data, pressure data, and geographic information data, wherein the heat data includes: the ambient temperature of the terminal and the temperature inside the pipeline, and the geographic information data includes: the geographic coordinates and floor location of the terminal node;

[0097] Establish a terminal data table, including: a data list for flow data, heat data, pressure data, and geographic information data, and establish connections of the same terminal group data between each data list according to the data labels of the node data in the data list.

[0098] In one embodiment of the present invention, constructing a similarity evaluation model between terminal nodes includes the following steps:

[0099] Obtain the terminal pipeline topology data, the flow data, heat data, pressure data, and geographical information data collected by the terminal nodes according to the geographical information data of the terminal; among them, the terminal pipeline topology data includes: the connecting pipelines and positions of the terminal nodes in the transmission pipeline network node management diagram, and the number of nodes directly connected by the terminal nodes;

[0100] Establish a terminal pipeline topology data table and add it to the established terminal data table;

[0101] Construct a comprehensive similarity recognition model for flow data and heat data through a deep learning algorithm;

[0102] Evaluate the comprehensive similarity of flow data and heat data between terminal nodes through the comprehensive similarity recognition model of flow data and heat data;

[0103] According to the comprehensive similarity evaluation result, the terminal pipeline topology data table, and the data tables of pressure data and geographical information data, establish a similarity evaluation model between terminal nodes through the following formula:

[0104]

[0105] Among them, S is the similarity evaluation value between terminal nodes, S1 is the similarity of terminal pipeline topology data between terminal nodes, S2 is the similarity of pressure data between terminal nodes, S3 is the similarity of geographical information data between terminal nodes, C is a constant set according to the comprehensive similarity evaluation result, w1, w2, and w3 are the weights of the similarity of terminal pipeline topology data between terminal nodes, the similarity of pressure data between terminal nodes, and the similarity of geographical information data between terminal nodes respectively; g0 is the mean value of the number of nodes directly connected by terminal nodes in the terminal group where the two terminal nodes are located, is the difference between the number of nodes directly connected by the two terminal nodes, Dg0 is the mean value of the distance between the terminal nodes and the connecting pipelines in the transmission pipeline network node management diagram in the terminal group where the two terminal nodes are located, Dg1 is the distance between the terminal node and the connecting pipelines in the transmission pipeline network node management diagram, Dg2 is the distance between the other terminal node and the connecting pipelines in the transmission pipeline network node management diagram, with the unit of km; S0 is the mean value of the variance of the terminal node pressure data in the preset time period in the terminal group where the two terminal nodes are located, is the difference between the variances of the terminal node pressure data in the preset time period, is the mean value of the pressure data of the terminal nodes in the terminal group where the two terminal nodes are located, with the unit of MPa, is the difference between the mean values of the terminal node pressure data in the preset time period; is the mean value of the distance between the geographical coordinates of the terminal nodes in the terminal group where the two terminal nodes are located, is the distance between the geographical coordinates of the terminal nodes, with the unit of km. is the sum of the total number of floors where the two terminal nodes are located. is the difference in the number of floors where the terminal nodes are located.

[0106] Specifically, in this embodiment, through summarizing a large amount of experimental data, w1 is set to 0.4, w2 is set to 0.4, w3 is set to 0.2, which are constants set according to the comprehensive similarity evaluation result. If the comprehensive similarity evaluation result is a high-similarity node, C is set to 2.5. If the comprehensive similarity evaluation result is a non-high-similarity node, C is set to 0.

[0107] In one embodiment of the present invention, a transmission pipeline network node similarity evaluation model is constructed, including the following steps:

[0108] Obtain the flow data, heat data, and sound data collected by the edge data nodes on the transmission pipeline network;

[0109] Evaluate the comprehensive similarity of the flow data and heat data between the terminal nodes through the comprehensive similarity recognition model of the flow data and heat data;

[0110] Extract the time-domain features and frequency-domain features from the audio data of the transmission pipeline network nodes with high similarity in the transmission pipeline network from historical data. Through the time-domain features and frequency-domain features, and through machine learning algorithms, train the sound data recognition model to identify whether the transmission pipeline network node is a high-similarity node;

[0111] Among them, extracting the time-domain features of the audio of the sound data of the pipeline includes: 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;

[0112] Identify the similarity of the transmission pipeline network node data through the comprehensive similarity recognition model of the flow data and heat data and the sound data recognition model: If the output data of the comprehensive similarity recognition model of the flow data and heat data, and the output data of the sound data recognition model both identify the data as high-similarity node data, then the transmission pipeline network node is a high-similarity node; otherwise, it is a non-high-similarity node.

[0113] In one embodiment of the present invention, through a deep learning algorithm, a comprehensive similarity recognition model of the flow data and heat data is constructed, including the following steps:

[0114] From historical data, for the flow data and heat data of the high-similarity nodes in the transmission pipeline network or terminal nodes, organize the data into a structured training data set and validation data set;

[0115] Using a training dataset and a validation dataset, a recurrent neural network is trained to establish a comprehensive similarity recognition model for flow data and heat data to identify whether the nodes of the transmission pipeline network are high-similarity nodes.

[0116] In one embodiment of the present invention, the similarity between terminal nodes evaluated by the similarity evaluation model between terminal nodes includes the following steps:

[0117] Using the similarity evaluation model between terminal nodes, calculate the similarity evaluation value between terminal nodes;

[0118] If the similarity evaluation value between terminal nodes is greater than the preset threshold, the terminal nodes are high-similarity terminals; otherwise, they are not high-similarity terminals.

[0119] Specifically, in this embodiment, if the similarity evaluation value between terminal nodes is greater than 5, the terminal nodes are high-similarity terminals; otherwise, they are not high-similarity terminals.

[0120] As an embodiment on the other hand of the present invention, a method for processing heating data of a heating system is provided, including the following steps:

[0121] Set edge data nodes at the transmission pipeline network and heat sources in the heating system. According to the grouping principle of grouping terminals into the same terminal group when they are connected to the same branch of the transmission pipeline network in the heating system, group the terminals in the heating system, and set data aggregation nodes on the same branch of the transmission pipeline network connected to the terminals in each group;

[0122] Set a data mirror warehouse and a wireless communication device at each edge data node, number the edge data nodes, set data tags for the data according to the time stamp of data collection and the node number, and sequentially establish data transmission channels between adjacent edge data nodes from the edge data node at the heat source to the data aggregation node;

[0123] The terminal data collected by each terminal in the terminal group is sent to the data aggregation node in the group. After the data aggregation node stores the data, it is packaged and sent to the nearest edge data node;

[0124] After the edge data node collects the data of this node, it stores the data of this node in the mirror warehouse, and packages the data of this node and the terminal data packet together and transmits them through the data transmission channel between the edge data nodes to the edge data node at the heat source;

[0125] The edge data node at the heat source standardizes the received data, establishes a data table for the node, and correlates the high-similarity data.

[0126] 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 heating data processing system for a heating system, characterized in that: include: Node setting module: edge data nodes are set at the transmission pipe network and heat source in the heating system. According to the grouping principle that terminals connected to the same transmission pipe network branch are grouped into the same terminal group, the terminals in the heating system are grouped, and data aggregation nodes are set on the same transmission pipe network branch connected to the terminals in each group; Transmission channel construction module: set up a data mirror warehouse and wireless communication equipment at each edge data node, number the edge data nodes, set data labels for the data according to the timestamp and node number of data collection, and establish data transmission channels between adjacent edge data nodes in sequence from the edge data node at the heat source to the data aggregation node; Terminal data aggregation module: The terminal data collected by each terminal in the terminal group is sent to the data aggregation node in the group. After the data aggregation node stores the data, it is packaged and sent to the nearest edge data node; Transmission management module: After the edge data node collects the data of the node, it stores the data of the node in the image warehouse, and packages the data of the node and the terminal data packet together and transmits them to the edge data node at the hot source through the data transmission channel between the edge data nodes; Data association module: The edge data node at the heat source standardizes the received data, establishes a node data table, evaluates the similarity between terminal nodes through the similarity evaluation model between terminal nodes, evaluates the similarity between transmission network nodes through the similarity evaluation model between transmission network nodes, and associates the highly similar data; Among them, building a similarity evaluation model between terminal nodes includes the following steps: Obtain terminal pipeline topology data, flow data, heat data, pressure data and geographic information data collected by the terminal node according to the terminal's geographic information data; wherein the terminal pipeline topology data includes: the connection pipeline and position of the terminal node in the transmission pipeline network node management diagram, and the number of nodes directly connected to the terminal node; Create a terminal pipeline topology data table and add it to the established terminal data table; Through deep learning algorithms, a comprehensive similarity recognition model for flow data and heat data is constructed; The comprehensive similarity recognition model of flow data and heat data is used to evaluate the comprehensive similarity of flow data and heat data between terminal nodes; Based on the comprehensive similarity evaluation results, the terminal pipeline topology data table, and the pressure data and geographic information data tables, a similarity evaluation model between terminal nodes is established using the following formula: Among them, S is the similarity evaluation value between terminal nodes, S1 is the similarity of terminal pipeline topology data between terminal nodes, S2 is the similarity of pressure data between terminal nodes, S3 is the similarity of geographic information data between terminal nodes, C is a constant set according to the comprehensive similarity evaluation result, w1, w2 and w3 are the weights of the similarity of terminal pipeline topology data between terminal nodes, the similarity of pressure data between terminal nodes and the similarity of geographic information data between terminal nodes respectively; g0 is the average number of nodes directly connected to the terminal nodes in the terminal group where the two terminal nodes are located, is the difference in the number of nodes directly connected to the two terminal nodes, Dg0 is the mean of the distances between the terminal nodes and the connecting pipelines in the transmission network node management diagram in the terminal group where the two terminal nodes are located, Dg1 is the distance between the terminal node and the connecting pipeline in the transmission network node management diagram, and Dg2 is the distance between the other terminal node and the connecting pipeline in the transmission network node management diagram; S0 is the mean of the variance of the terminal node pressure data in the terminal group where the two terminal nodes are located within a preset time period, is the difference in variance of the terminal node pressure data within the preset time period, is the mean value of the terminal node pressure data in the terminal group where the two terminal nodes are located, It is the difference between the mean values ​​of the terminal node pressure data within the preset time period; is the mean distance between the geographical coordinates of the two terminal nodes in the terminal group where the two terminal nodes are located, is the distance between the geographical coordinates of the terminal nodes, is the sum of the total number of floors where the two terminal nodes are located, is the difference in the number of floors where the terminal nodes are located; Constructing a transmission network node similarity evaluation model includes the following steps: Obtain flow data, heat data and sound data collected by edge data nodes on the transmission pipeline network; The comprehensive similarity recognition model of flow data and heat data is used to evaluate the comprehensive similarity of flow data and heat data between terminal nodes; Through the audio data of the transmission network nodes with high similarity in the transmission network in the historical data, the time domain features and frequency domain features are extracted. Through the time domain features and frequency domain features, the sound data recognition model is trained through the machine learning algorithm to identify whether the transmission network node is a high similarity node; The time domain features of the sound data audio of the pipeline are extracted, including mean, variance and peak value, and the time domain signal of the audio is converted into frequency domain representation through Fourier transform to extract frequency domain features, including frequency width and power spectrum density; The similarity of the transmission network node data is identified through the comprehensive similarity recognition model of flow data and heat data and the sound data recognition model: if the output data of the comprehensive similarity recognition model of flow data and heat data, and the output data of the sound data recognition model, are both identified as high-similarity node data, then the transmission network node is a high-similarity node; otherwise, it is a non-high-similarity node.

2. The heating data processing system of a heating system according to claim 1, characterized in that: Setting up a data mirror warehouse and wireless communication equipment at each edge data node and numbering the edge data nodes includes the following steps: A wireless communication device and a data mirror warehouse are set up at each edge data node, wherein the data mirror warehouse is used to store copies of data collected by the edge node; Assign a unique identifier to each edge data node as the edge data node number, create a transmission network node management map, and record the transmission network location of each edge node, the node number, and the number of nodes with the least distance between the edge data node at the heat source.

3. The heating data processing system of a heating system according to claim 1, characterized in that: From the edge data node at the heat source to the data aggregation node, a data transmission channel between adjacent edge data nodes is established in sequence, including the following steps: From the edge data node at the heat source to the data aggregation node, according to the transmission network node management diagram, the data transmission channel between adjacent edge data nodes is established in sequence through the wireless communication equipment on the transmission network; The aggregation node establishes a data transmission channel with the nearest edge data node through the MQTT communication protocol.

4. The heating data processing system of a heating system according to claim 1, characterized in that: The edge data node at the heat source standardizes the received data, establishes a node data table, and associates the highly similar data, including the following steps: The edge data node at the heat source aggregates the data sent by the edge data nodes and sends it together with the data of the node to the cloud management platform. The cloud management platform standardizes the data of each edge data node and establishes the terminal data table and the data table of the transmission pipeline network node; Based on the data collected by the heating system terminal and the node data collected by the edge data nodes on the transmission pipeline network, a similarity evaluation model between terminal nodes and a similarity evaluation model for transmission pipeline network nodes are constructed; The similarity between terminal nodes evaluated by the similarity evaluation model between terminal nodes, and the similarity between transmission network nodes evaluated by the transmission network node similarity evaluation model; The data of terminals with high similarity are associated with each other, and the data of transmission network nodes with high similarity are associated with each other.

5. The heating data processing system of a heating system according to claim 4, characterized in that: The cloud management platform standardizes the data of each edge data node and establishes the terminal data table and the data table of the transmission network node, including the following steps: Establish data standardization rules for edge data nodes to convert the data collected by each edge data node into the same format; Node data collected by edge data nodes include: flow data, heat data and sound data, wherein the heat data includes: internal temperature data of the pipeline and external temperature data of the pipeline, and the sound data includes external sound audio data of the pipeline and internal sound audio data of the pipeline; The edge data nodes at the heat source collect node data such as flow data, heat data and sound data at the heat source outlet and return water outlet; Establish data tables for transmission network nodes, including data lists for monitoring flow data, heat data, and sound data, and establish connections between the same node data in each data list based on the data labels of the node data in the data list; Set up terminal nodes at the heating system terminals to collect data, number the terminal nodes, and add data labels to the data of the terminal nodes according to the terminal node numbers and terminal groups; The data collected by the terminal node include: flow data, heat data, pressure data and geographic information data, wherein the heat data includes: the ambient temperature of the terminal and the temperature in the pipeline, and the geographic information data includes: the geographic coordinates and floor location of the terminal node; Establish a terminal data table, including: data lists of flow data, heat data, pressure data and geographic information data, and establish connections between the same terminal grouping data between various data lists based on the data labels of the node data in the data lists.

6. The heating data processing system of a heating system according to claim 1, characterized in that: Through the deep learning algorithm, a comprehensive similarity recognition model for flow data and heat data is constructed, including the following steps: Through the flow data and heat data of high-similarity nodes in the historical data, transmission pipeline network or terminal nodes, the data is organized into structured training data sets and verification data sets; Through the training data set and the verification data set, the recurrent neural network is trained, and a comprehensive similarity recognition model of flow data and heat data is established to identify whether the transmission pipeline network node is a high-similarity node.

7. The heating data processing system of a heating system according to claim 4, characterized in that: The similarity between terminal nodes evaluated by the similarity evaluation model between terminal nodes includes the following steps: Calculate the similarity evaluation value between terminal nodes through the similarity evaluation model between terminal nodes; If the similarity evaluation value between the terminal nodes is greater than a preset threshold, the terminal node is a high-similarity terminal, otherwise, it is not a high-similarity terminal.

8. A method for processing heating data of a heating system, characterized in that: Applied in the system as claimed in any one of claims 1 to 7, comprising the following steps: Edge data nodes are set at the transmission pipe network and heat source in the heating system. According to the grouping principle that terminals connected to the same transmission pipe network branch are grouped into the same terminal group, the terminals in the heating system are grouped, and data aggregation nodes are set on the same transmission pipe network branch connected to the terminals in each group; Set up a data mirror warehouse and wireless communication equipment at each edge data node, number the edge data nodes, set data labels for the data according to the timestamp and node number of the data collection, and establish data transmission channels between adjacent edge data nodes in sequence from the edge data node at the heat source to the data aggregation node; The terminal data collected by each terminal in the terminal group is sent to the data aggregation node in the group. After the data aggregation node stores the data, it is packaged and sent to the nearest edge data node; After the edge data node collects the data of the node, it stores the data of the node in the image warehouse, and packages the data of the node and the terminal data packet together and transmits them to the edge data node at the hot source through the data transmission channel between the edge data nodes; The edge data node at the heat source standardizes the received data, establishes the node data table, and associates the highly similar data.

Images