Neural network driven heat supply network fault analysis method and system

Abstract

The invention provides a neural network driven heat supply network fault analysis method and system, and the method comprises the steps: obtaining the current operation state data and current control parameter variation of a heat source, the current operation state data and current control parameter variation of a heat supply network, and fault judgment conditions, and determining a control data combination; if the control data combination is not matched with the heat supply network abnormal control parameter range combination and the current operation state data of the heat supply network is not matched with the heat supply network abnormal state data range, inputting the current operation state data of the heat source, the current control parameter variable quantity of the heat source and the current control parameter variable quantity of the heat supply network into a heat supply network operation state output model; the operation state data variable quantity of the heat supply network and the updating state data of the heat supply network are obtained, and a heat supply network fault analysis result is determined according to the matching result of the updating state data of the heat supply network and the heat supply network abnormal state data range; the fault analysis result of the heat supply network is determined in time, and the probability of abnormal operation of the heat supply network caused by potential faults or dangerous factors is reduced.

Description

Technical Field

[0001] This application relates to the field of centralized heating technology, and in particular to a neural network-driven method and system for analyzing heating network faults. Background Technology

[0002] A heating network (also known as a heat supply network) is a core component of a centralized heating system. It is a network of pipes that connects heat sources and heat users (such as residential buildings, commercial buildings, and industrial facilities), and is responsible for efficiently and stably delivering heat energy to end users in the form of hot water or steam.

[0003] The heating network mainly includes: the primary network (the transmission network between the heat source and the heat exchange station), the secondary network (the transmission network between the heat exchange station and the user's radiators or underfloor heating system), and the heat exchange station itself. The stable operation of the heating network directly affects the heating experience of end users and the system's energy efficiency. Therefore, relevant technologies involve real-time monitoring of the heating network's operation to enable timely responses to faults.

[0004] However, the current technology relies on real-time monitoring of the heating network's operation to determine whether a fault has occurred. This method is inherently delayed and cannot prevent potential faults or existing hazards that could cause abnormal network operation. Therefore, improvements to the heating network fault monitoring methods in these technologies are necessary. Summary of the Invention

[0005] In view of the shortcomings of the prior art described above, this application provides a neural network-driven method and system for analyzing thermal network faults to solve the above-mentioned technical problems.

[0006] According to one aspect of the embodiments of this application, a neural network-driven method for analyzing heating network faults is provided. The method includes: acquiring current operating status data and current control parameter changes of a heat source, current operating status data of the heating network, current control parameter changes, and fault determination conditions; the fault determination conditions include: a range of abnormal heating network status data and a combination of abnormal heating network control parameter ranges; the current control parameter changes of the heat source are determined by the control parameter data of the heat source at the previous moment and the current control parameter data; the current control parameter changes of the heating network are determined by the control parameter data of the heating network at the previous moment and the current control parameter data; combining the current control parameter data of the heat source and the current control parameter data of the heating network to obtain a control data combination; if If the combination of control data does not match the range of abnormal control parameters for the heating network, and the current operating status data of the heating network does not match the range of abnormal state data for the heating network, then the current operating status data of the heat source, the change in the current control parameters of the heat source, and the change in the current control parameters of the heating network are input into the heating network operating status output model to obtain the change in the operating status data of the heating network. The heating network operating status output model is trained on a pre-constructed operating status output model based on sample data. The sum of the current operating status data of the heating network and the change in the operating status data of the heating network is used as the updated status data of the heating network. Based on the matching result between the updated status data of the heating network and the range of abnormal state data of the heating network, the heating network fault analysis result is determined.

[0007] In one embodiment of this application, if the pre-built operating state output model includes: an input layer, a residual module, a pooling layer, and a fully connected layer, and the sample data includes: historical operating state data and historical control parameter changes of the heat source, and historical operating state data and historical control parameter changes of the heating network, then the process of training the pre-built operating state output model based on the sample data to obtain the heating network operating state output model includes: inputting the historical operating state data of the heat source and the historical operating state data of the heating network into the input layer to obtain the historical operating state data changes of the heat source and the historical operating state data changes of the heating network; inputting the historical operating state data changes of the heat source, the historical control parameter changes of the heat source, the historical operating state data changes of the heating network, and the historical control parameter changes of the heating network into the residual module. The difference module obtains data association features, which include: association functions between the control parameters of the heat source and the operating status data of the heat source, association functions between the operating status data of the heat source and the operating status data of the heating network, and association functions between the control parameters of the heating network and the operating status data of the heating network. These data association features are then input into the pooling layer to obtain data association features with reduced dimensionality. The reduced-dimensional data association features are then input into the fully connected layer to obtain data association functions. These data association functions include: association functions between the control parameters of the heat source, the operating status data of the heat source, the operating status data of the heating network, and the control parameters of the heating network. By verifying the data and the data association functions, the parameters in the pre-constructed operating status output model are adjusted to obtain the heating network operating status output model.

[0008] In one embodiment of this application, the process of determining the heating network fault analysis result based on the matching result of the updated status data of the heating network and the range of abnormal status data of the heating network includes: if the updated status data of the heating network is within the range of abnormal status data of the heating network, then it is determined that the updated status data of the heating network matches the range of abnormal status data of the heating network, and the existence of fault risk is taken as the heating network fault analysis result; if the updated status data of the heating network is not within the range of abnormal status data of the heating network, then it is determined that the updated status data of the heating network does not match the range of abnormal status data of the heating network, and the absence of fault risk is taken as the heating network fault analysis result.

[0009] In one embodiment of this application, after obtaining the failure analysis results of a heating network with potential failure risk, the method further includes: obtaining the control parameter combination of the heating network under no failure risk conditions and the control parameter combination of the heat source under normal operation conditions; merging the control parameter combination of the heating network under no failure risk conditions and the control parameter combination of the heat source under normal operation conditions to obtain the normal control parameter combination of the heating network; determining the heat generated by the heating network within a preset time period based on the normal control parameter combination of the heating network, denoted as the first heat; determining the production cost of the first heat based on the first heat; determining the evaluation parameters of the normal control parameter combination of the heating network based on the first heat and the production cost of the first heat; filtering the normal control parameter combination based on the evaluation parameters of the normal control parameter combination to obtain a first filtered control parameter combination; and determining a target control parameter combination based on the evaluation parameters of the first filtered control parameter combination and the difference between the control data in the first filtered control parameter combination and the control data in the control data combination.

[0010] In one embodiment of this application, the evaluation parameters for the normal control parameter combination of the heating network include: ,in, Evaluation parameters representing the normal control parameter combination of the heating network. This indicates the heating capacity of the centralized heating system, determined by the heat generated within a preset time period. This represents the adjustment coefficient. The heat cost rate of a centralized heating system is determined by the heat generated within a preset time period and the production cost of the heat generated within the preset time period.

[0011] In one embodiment of this application, the process of determining a target control parameter combination based on the evaluation parameters of the first screening control parameter combination and the difference between the control data in the first screening control parameter combination and the control data in the control data combination includes: recording the difference between the control data in the first screening control parameter combination and the control data in the control data combination as a first difference; if the first difference is greater than or equal to a preset difference threshold, updating the first screening control parameter combination, and determining the target control parameter combination based on the updated control parameter combination; if the first difference is less than the preset difference threshold, using the first screening control parameter combination for which the first difference is calculated as a second screening control parameter combination; if the number of second screening control parameter combinations is greater than a preset number threshold, using the second screening control parameter combination with the largest evaluation parameter as the target control parameter combination. If the number of the second filtering control parameter combinations is less than or equal to the preset number threshold, then the second filtering control parameter combination is used as the target control parameter combination.

[0012] In one embodiment of this application, the process of updating the first screening control parameter combination and determining the target control parameter combination based on the updated control parameter combination includes: using the first screening control parameter combination as a population individual and the evaluation parameter of the first screening control parameter combination as the fitness of the population individual; performing crossover and mutation on the population individuals to obtain updated population individuals; screening the updated population individuals based on their fitness to obtain screened population individuals; and determining the target control parameter combination based on the fitness of the screened population individuals and the difference between the control data in the screened population individuals and the control data in the control data combination.

[0013] In one embodiment of this application, the formula for calculating the first difference includes: , in, Indicates the first difference, Indicates the first Weighting coefficients for control data items Indicates the first in the first combination of screening control parameters Item control data, Indicates the first in the control data combination Item control data.

[0014] In one embodiment of this application, before acquiring the current operating status data and current control parameter changes of the heat source, the current operating status data, current control parameter changes, and fault judgment conditions of the heating network, the method further includes: collecting the current heat supply, current outlet water supply temperature, current outlet water supply pressure, current outlet water supply flow rate, current inlet return water temperature, current fuel consumption control value, current equipment load rate control value, and current supply and return water temperature difference control value during the operation of the heat source; and collecting the current outlet water supply temperature, current outlet water supply pressure, current outlet water supply flow rate, current inlet return water temperature, current circulating pump operating frequency control value, current flow valve opening control value, and current supply and return water temperature difference control value during the operation of the heating network; and combining the current heat supply, current outlet water supply temperature, current outlet water supply pressure, current outlet water supply flow rate, and current inlet return water temperature during the operation of the heat source. The following parameters are used as the current operating status data of the heat source: current fuel consumption control value, current equipment load rate control value, and current supply and return water temperature difference control value during the operation of the heat source; current outlet water supply temperature, current outlet water supply pressure, current outlet water supply flow rate, and current inlet return water temperature during the operation of the heating network; current circulating pump operating frequency control value, current flow valve opening control value, and current supply and return water temperature difference control value during the operation of the heating network; when the heat source or the heating network malfunctions, the current control parameter data of the heat source and the current control parameter data of the heating network are combined to obtain the abnormal control parameter range combination of the heating network; and the abnormal range of the current operating status data of the heating network is used as the abnormal status data range of the heating network.

[0015] According to one aspect of the embodiments of this application, a neural network-driven heating network fault analysis system is provided, comprising: a data acquisition module, configured to acquire current operating status data and current control parameter changes of a heat source, current operating status data of the heating network, current control parameter changes, and fault determination conditions; the fault determination conditions include: a range of abnormal heating network status data and a combination of abnormal heating network control parameter ranges; the current control parameter changes of the heat source are determined by the control parameter data of the heat source at the previous moment and the current control parameter data; the current control parameter changes of the heating network are determined by the control parameter data of the heating network at the previous moment and the current control parameter data; a data combination module, configured to combine the current control parameter data of the heat source and the current control parameter data of the heating network to obtain a control data combination; the change amount is determined by... The module is configured to, if the control data combination does not match the range of abnormal control parameters for the heating network, and the current operating status data of the heating network does not match the range of abnormal state data for the heating network, input the current operating status data of the heat source, the change in the current control parameters of the heat source, and the change in the current control parameters of the heating network into the heating network operating status output model to obtain the change in the operating status data of the heating network; the heating network operating status output model is obtained by training a pre-constructed operating status output model based on sample data; the fault analysis module is configured to take the sum of the current operating status data of the heating network and the change in the operating status data of the heating network as the updated status data of the heating network; and determine the fault analysis result of the heating network based on the matching result of the updated status data of the heating network and the range of abnormal state data of the heating network.

[0016] The beneficial effects of this application are as follows: This application obtains the current operating status data and current control parameter changes of the heat source, the current operating status data, current control parameter changes, and fault judgment conditions of the heating network. It then combines the current control parameter data of the heat source and the heating network to obtain a control data combination. If the control data combination does not match the abnormal control parameter range combination of the heating network, and the current operating status data of the heating network does not match the abnormal status data range of the heating network, then the current operating status data of the heat source, the current control parameter changes of the heat source, and the current control parameter changes of the heating network are input into the heating network operating status output model to obtain the operating status data changes of the heating network. Finally, the current operating status data of the heating network and the operating status data of the heating network are combined... The sum of changes is used as the updated status data of the heating network. Based on the matching results of the updated status data of the heating network with the range of abnormal status data of the heating network, the heating network fault analysis results are determined. In the above process, when the current control parameters of the heat source change, and / or the current control parameters of the heating network change, and under the condition that the operating status of the heating network and the operating status of the heat source are both normal, the changes in the current operating status data of the heating network caused by the changes in the current control parameters of the heat source and / or the current control parameters of the heating network are promptly known. Thus, based on the matching results of the updated status data of the heating network with the range of abnormal status data of the heating network, the heating network fault analysis results are promptly determined, and the probability of potential fault risks or dangerous factors leading to abnormal operation of the heating network is reduced.

[0017] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. In the drawings: Figure 1 This is a schematic diagram illustrating an exemplary system architecture as shown in an exemplary embodiment of this application; Figure 2 This is a flowchart illustrating a neural network-driven heating network fault analysis method in an exemplary embodiment of this application; Figure 3 This is a schematic diagram illustrating the structure of a heating network operation status output model, as shown in an exemplary embodiment of this application. Figure 4 This is a block diagram illustrating a neural network-driven thermal network fault analysis system, as shown in an exemplary embodiment of this application. Detailed Implementation

[0019] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0020] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. The drawings only show the components related to this application and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0021] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the present application. However, it will be apparent to those skilled in the art that embodiments of the present application may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the present application.

[0022] Figure 1 This is a schematic diagram illustrating an exemplary system architecture as shown in an exemplary embodiment of this application.

[0023] Reference Figure 1As shown, the system architecture may include a data acquisition device 101 and a fault analysis device 102. The fault analysis device 102 may be at least one of a desktop graphics processing unit (GPU) computer, a GPU computing cluster, a neural network computer, etc. Relevant technical personnel can use the fault analysis device 102 to acquire the current operating status data and current control parameter changes of the heat source, the current operating status data and current control parameter changes of the heating network, and fault judgment conditions. The current control parameter data of the heat source and the heating network are combined to obtain a control data combination. If the control data combination does not match the heating network abnormal control parameter range combination, and the current operating status data of the heating network does not match the heating network abnormal status data range, then the current operating status data of the heat source, the current control parameter changes of the heat source, and the current control parameter changes of the heating network are input into the heating network operating status output model to obtain the heating network operating status data changes. The sum of the current operating status data of the heating network and the heating network operating status data changes is used as the updated status data of the heating network. Based on the matching result between the updated status data of the heating network and the heating network abnormal status data range, the heating network fault analysis result is determined. The data acquisition device 101 is used to acquire the current operating status data and current control parameter changes of the heat source, the current operating status data, current control parameter changes and fault judgment conditions of the heating network, and provide them to the fault analysis device 102 for processing.

[0024] Schematic, after acquiring the current operating status data and current control parameter changes from the acquisition device 101, the fault analysis device 102 combines the current control parameter data of the heat source and the current control parameter changes of the heat network to obtain a control data combination. If the control data combination does not match the abnormal control parameter range combination of the heat network, and the current operating status data of the heat network does not match the abnormal status data range of the heat network, then the current operating status data of the heat source, the current control parameter changes of the heat source, and the current control parameter changes of the heat network are input into the heat network operating status output model to obtain the operating status data changes of the heat network. The sum of changes in the status data is used as the updated status data of the heating network. Based on the matching results between the updated status data and the range of abnormal status data of the heating network, the heating network fault analysis results are determined. In the above process, when the current control parameters of the heat source change, and / or the current control parameters of the heating network change, and under the condition that the operating status of the heating network and the operating status of the heat source are both normal, the changes in the current operating status data of the heating network caused by the changes in the current control parameters of the heat source and / or the current control parameters of the heating network are promptly known. Thus, based on the matching results between the updated status data and the range of abnormal status data of the heating network, the heating network fault analysis results are promptly determined, and the probability of potential fault risks or dangerous factors leading to abnormal operation of the heating network is reduced.

[0025] The implementation details of the technical solutions in the embodiments of this application are described in detail below: Figure 2 This is a flowchart illustrating a neural network-driven thermal network fault analysis method according to an exemplary embodiment of this application. (Refer to...) Figure 2 As shown, the neural network-driven heating network fault analysis method includes at least steps S210 to S240, which are described in detail below: In step S210, the current operating status data and current control parameter changes of the heat source, the current operating status data and current control parameter changes of the heating network, and fault judgment conditions are acquired. In one embodiment of this application, the current operating status data of the heat source includes: current heat supply, current outlet water supply temperature, current outlet water supply pressure, current outlet water supply flow rate, and current inlet return water temperature, etc., during the operation of the heat source. The current control parameter data of the heat source includes: current fuel consumption control value, current equipment load rate control value, and current supply and return water temperature difference control value, etc., during the operation of the heat source. The current operating status data of the heating network includes: current outlet water supply temperature, current outlet water supply pressure, current outlet water supply flow rate, and current inlet return water temperature, etc., during the operation of the heating network. The current control parameter data of the heating network includes: current circulating pump operating frequency control value, current flow valve opening control value, and current supply and return water temperature difference control value, etc., during the operation of the heating network. Fault determination criteria include: the range of abnormal heating network status data and the combination of abnormal heating network control parameter ranges. The abnormal heating network status data range includes: the abnormal range of current heating output, the abnormal range of current outlet water supply temperature, the abnormal range of current outlet water supply pressure, the abnormal range of current outlet water supply flow rate, and the abnormal range of current inlet return water temperature. The abnormal heating network status data range needs to be set according to the design requirements of the heating network. The combination of abnormal heating network control parameter ranges includes: the abnormal control range of heat source fuel consumption, the abnormal control range of heat source equipment load rate, the abnormal control range of heat source supply and return water temperature difference, the abnormal control range of heating network circulating pump operating frequency, the abnormal control range of heating network flow valve opening, and the abnormal control range of heating network supply and return water temperature difference. The abnormal control range of heat source fuel consumption consists of the fuel consumption when the heat source or the heating network experiences a fault; the abnormal control range of heat source equipment load rate consists of... The equipment load rate is composed of the following parameters when the heat source or heating network fails: the supply and return water temperature difference control range is composed of the supply and return water temperature difference when the heat source or heating network fails; the circulating pump operating frequency control range is composed of the circulating pump operating frequency when the heat source or heating network fails; the flow valve opening degree control range is composed of the flow valve opening degree when the heat source or heating network fails; and the supply and return water temperature difference control range is composed of the supply and return water temperature difference when the heat source or heating network fails.The change in the current control parameters of the heat source is determined by the control parameter data of the heat source at the previous moment and the current control parameter data. That is, the change in the current control parameters of the heat source is the difference between the current control parameter data and the control parameter data of the heat source at the previous moment. If the control parameter data of the heat source includes multiple items, the sum of the absolute values ​​of the differences between the current and previous moments for each control parameter is taken as the change in the current control parameters of the heat source. The change in the current control parameters of the heating network is determined by the control parameter data of the heating network at the previous moment and the current control parameter data. That is, the change in the current control parameters of the heating network is the difference between the current and previous moments for the heating network. If the control parameter data of the heating network includes multiple items, the sum of the absolute values ​​of the differences between the current and previous moments for each control parameter is taken as the change in the current control parameters of the heating network.

[0026] In step S220, the current control parameter data of the heat source and the current control parameter data of the heating network are combined to obtain a control data combination. In one embodiment of this application, the process of combining the current control parameter data of the heat source and the current control parameter data of the heating network is as follows: the current control parameter data of the heat source and the current control parameter data of the heating network are combined sequentially, and the format of the control data combination is [current control parameter name 1 of heat source: data of current control parameter name 1 of heat source; current control parameter name 2 of heat source: data of current control parameter name 2 of heat source; current control parameter name 3 of heat source: data of current control parameter name 3 of heat source; current control parameter name 1 of heating network: data of current control parameter name 1 of heating network...].

[0027] In step S230, if the control data combination does not match the range of abnormal control parameters for the heating network, and the current operating status data of the heating network does not match the range of abnormal state data for the heating network, then the current operating status data of the heat source, the change in the current control parameters of the heat source, and the change in the current control parameters of the heating network are input into the heating network operating status output model to obtain the change in the operating status data of the heating network. In one embodiment of this application, the heating network operating status output model is obtained by training a pre-constructed operating status output model based on sample data. The process of matching control data combinations with the range of abnormal control parameters in the heating network includes: comparing each control parameter in the control data combination with the corresponding abnormal control parameter range in the range of abnormal control parameters in the heating network; if each control parameter in the control data combination falls within the corresponding abnormal control parameter range in the range of abnormal control parameters in the range of abnormal control parameters in the heating network, then the control data combination is determined to match the range of abnormal control parameters in the heating network; if at least one control parameter in the control data combination does not fall within the corresponding abnormal control parameter range in the range of abnormal control parameters in the range of abnormal control parameters in the heating network, then the control data combination is determined to not match the range of abnormal control parameters in the heating network. The process of matching the current operating status data of the heating network with the range of abnormal status data in the heating network includes: comparing the current operating status data of the heating network with the range of abnormal status data in the heating network; if the current operating status data of the heating network falls within the range of abnormal status data in the heating network, then the current operating status data of the heating network is determined to match the range of abnormal status data in the heating network; if the current operating status data of the heating network does not fall within the range of abnormal status data in the heating network, then the current operating status data of the heating network is determined to not match the range of abnormal status data in the heating network. If the current operating status data of the heating network contains multiple types of operating status data, then each type of operating status data is compared with the corresponding type of abnormal status data range in the heating network abnormal status data range. If any type of operating status data falls within the corresponding type of abnormal status data range, then the current operating status data of the heating network is determined to match the heating network abnormal status data range. If none of the operating status data types fall within the corresponding type of abnormal status data range, then the current operating status data of the heating network is determined to not match the heating network abnormal status data range. If the control data combination does not match the heating network abnormal control parameter range combination, and the current operating status data of the heating network does not match the heating network abnormal status data range, then the operating status of the heating network is considered normal.

[0028] In one embodiment of this application, the determination that the control data combination does not match the range of abnormal control parameters of the heating network, and that the current operating status data of the heating network does not match the range of abnormal status data of the heating network, is performed under the condition that the heat source and the radiator are operating normally. Whether the operating status of the heating network is normal is determined by whether the operating status data of the heating network is within the normal range, and whether the operating status of the radiator is normal is determined by whether the operating status data of the radiator is within the normal range.

[0029] In one embodiment of this application, the normal operation of the heating network is determined by judging whether the combination of control data matches the combination of abnormal control parameters of the heating network. This avoids the lag caused by comparing real-time operating status data with the range of abnormal operating status data and reduces the probability of potential faults occurring during the operation of the heating network.

[0030] In step S240, the sum of the current operating status data and the changes in the operating status data of the heating network is used as the updated status data of the heating network. Based on the matching result between the updated status data and the range of abnormal status data of the heating network, the heating network fault analysis result is determined. In one embodiment of this application, when the current control parameters of the heat source change, and / or the current control parameters of the heating network change, and under the condition that both the operating status of the heating network and the operating status of the heat source are normal, the change in the current operating status data of the heating network caused by the change in the current control parameters of the heat source and / or the change in the current control parameters of the heating network is promptly known. Therefore, based on the matching result between the updated status data and the range of abnormal status data of the heating network, the heating network fault analysis result is determined promptly, and the probability of potential fault risks or dangerous factors leading to abnormal heating network operation is reduced.

[0031] In one embodiment of this application, if the pre-built operating status output model includes: an input layer, a residual module, a pooling layer, and a fully connected layer, and the sample data includes: historical operating status data and historical control parameter changes of the heat source, and historical operating status data and historical control parameter changes of the heating network, then the process of training the pre-built operating status output model based on the sample data to obtain the heating network operating status output model includes: Historical operating status data of the heat source and the heating network are input into the input layer to obtain the changes in historical operating status data of the heat source and the heating network. In one embodiment of this application, the changes in historical operating status data of the heat source include the changes in operating status data of the heat source at multiple different historical moments, and the changes in operating status data of the heat source at each historical moment are the differences between the operating status data at that historical moment and the operating status data at the previous historical moment. If the operating status data of the heat source at that historical moment includes multiple items, the sum of the absolute values ​​of the differences between each item of operating status data at that historical moment and the previous historical moment is taken as the changes in operating status data of the heat source at that historical moment. The changes in historical operating status data of the heating network include the changes in operating status data of the heating network at multiple different historical moments, and the changes in operating status data of the heating network at each historical moment are the differences between the operating status data at that historical moment and the operating status data at the previous historical moment. If the operating status data of the heating network at this historical moment includes multiple items, then the sum of the absolute values ​​of the differences between each operating status data item at this historical moment and the previous historical moment is taken as the change in the operating status data of the heating network at this historical moment.

[0032] The historical operational status data changes of the heat source, the historical control parameter changes of the heat source, the historical operational status data changes of the heating network, and the historical control parameter changes of the heating network are input into the residual module to obtain data association characteristics. In one embodiment of this application, the data association characteristics include: association functions between the control parameters of the heat source and the operational status data of the heat source, association functions between the operational status data of the heat source and the operational status data of the heating network, and association functions between the control parameters of the heating network and the operational status data of the heating network. The expression for the association function between the control parameters of the heat source and the operational status data of the heat source is as follows: Equation (1) in, This represents the correlation function between the control parameters of the heat source and the operating status data of the heat source. This represents the Sigmoid activation function. This represents the weight matrix of the second-level linear transformation. Indicates a corrected linear unit. This represents the weight matrix of the first-level linear transformation. This represents the historical changes in the control parameters of the heat source. This represents the change in the historical operating status data of the heat source. This represents the bias vector of the first layer of the neural network. This represents the bias vector of the second layer of the neural network.

[0033] The correlation function expression between the operating status data of the heat source and the operating status data of the heating network is shown below: Equation (2) in, This represents the correlation function between the operating status data of the heat source and the operating status data of the heating network. Represents the normalized exponential function, This represents the change in the historical operating status data of the heat source. This indicates the changes in the historical operating status data of the heating network.

[0034] Equation (3) in, This represents the correlation function between the control parameters of the heating network and the operating status data of the heating network. Represents the hyperbolic tangent function. Indicates the first Each weighting coefficient This represents the total number of weighting coefficients. The first [item] of the hot network The change in each historical control parameter The first [item] of the hot network The historical operating status data changes. When the number of historical control parameters of the heating network is inconsistent with the number of historical operating status data of the heating network, the insufficient data bits are filled with 0 to make the number of bits of the historical control parameters of the heating network the same as the number of bits of the historical operating status data of the heating network, so that the number of bits of the historical control parameter changes of the heating network is the same as the number of bits of the historical operating status data changes of the heating network.

[0035] In one embodiment of this application, the residual module includes at least two convolutional layers, namely, after the first convolutional layer, a normalization function and an activation function are set, and after the second convolutional layer, a normalization function is set.

[0036] The data association features are input into the pooling layer to obtain the data association features with reduced dimensionality. In one embodiment of this application, the pooling layer is used to reduce the number of parameters that need to be processed in the runtime output model, thereby improving the computational efficiency of the runtime output model.

[0037] The reduced-dimensional data association features are input into a fully connected layer to obtain a data association function. In one embodiment of this application, the data association function includes: an association function between the control parameters of the heat source, the operating status data of the heat source, the operating status data of the heating network, and the control parameters of the heating network; the calculation formula for the association function between the control parameters of the heat source, the operating status data of the heat source, the operating status data of the heating network, and the control parameters of the heating network is as follows: Equation (4) in, This represents the correlation function between the control parameters of the heat source, the operating status data of the heat source, the operating status data of the heating network, and the control parameters of the heating network. This represents the correlation function between the control parameters of the heat source and the operating status data of the heat source. This represents the correlation function between the operating status data of the heat source and the operating status data of the heating network. This represents the correlation function between the control parameters of the heating network and the operating status data of the heating network. This represents the activation function, which can be either the sigmoid activation function or the hyperbolic tangent function. This represents the weight matrix of the fully connected layer. This represents the bias vector of the fully connected layer.

[0038] By verifying data and using a data association function, the parameters in a pre-built operational transition output model are adjusted to obtain the heating network operational status output model. In one embodiment of this application, the verification data includes: changes in the verified operational status data of the heat source, changes in the verified control parameters of the heat source, changes in the verified operational status data of the heating network, and changes in the verified control parameters of the heating network. The process of adjusting the parameters in the pre-built operational transition output model using the verification data and the data association function to obtain the heating network operational status output model includes: inputting the changes in the verified operational status data of the heat source, the changes in the verified control parameters of the heat source, and the changes in the verified control parameters of the heating network into the data association function to obtain the changes in the output operational status data of the heating network; if the changes in the output operational status data of the heating network are consistent with the changes in the heating network's operational status data... If the deviation between the change in the verified operating status data and the change in the verified operating status data of the heating network is less than a preset deviation threshold, then the pre-built operating status output model is used as the operating status output model of the heating network. If the deviation between the change in the output operating status data of the heating network and the change in the verified operating status data of the heating network is greater than or equal to the preset deviation threshold, then the parameters in the pre-built operating status output model are adjusted until the deviation between the change in the output operating status data of the heating network obtained through the adjusted operating status output model and the change in the verified operating status data of the heating network is less than the preset deviation threshold. Then the adjusted operating status output model is used as the operating status output model of the heating network.

[0039] In one embodiment of this application, if the verification operation status data of the heating network includes multiple items, then the deviation between the output operation status data and the verification operation status data of the heating network being less than a preset deviation threshold means that the deviation between each item in the output operation status data and the corresponding item in the verification operation status data of the heating network is less than the preset deviation threshold. The preset deviation threshold can be set separately for different items of operation status data, or the same value can be set for different items of operation status data. By training the pre-built operation status output model to obtain the heating network operation status output model, and then training and verifying its accuracy, the accuracy of the heating network operation status output model in predicting the changes in the operation status data of the heating network is improved.

[0040] In one embodiment of this application, the process of determining the heating network fault analysis result based on the matching result of the updated status data of the heating network and the range of abnormal status data of the heating network includes: If the updated status data of the heating network falls within the range of abnormal status data, it is determined that the updated status data matches the range of abnormal status data, and the existence of a fault risk is taken as the result of the heating network fault analysis. In one embodiment of this application, the range of abnormal status data includes: the abnormal range of current heating capacity, the abnormal range of current outlet water supply temperature, the abnormal range of current outlet water supply pressure, the abnormal range of current outlet water supply flow rate, and the abnormal range of current inlet return water temperature, etc. The range of abnormal status data needs to be set according to the design requirements of the heating network. If the updated status data of the heating network falls within the range of abnormal status data, it means that at least one item in the updated status data of the heating network falls within the abnormal range of the corresponding item.

[0041] If the updated status data of the heating network is not within the range of abnormal status data, it is determined that the updated status data of the heating network does not match the range of abnormal status data, and the absence of fault risk is taken as the result of the heating network fault analysis. In one embodiment of this application, the updated status data of the heating network being outside the range of abnormal status data means that all items in the updated status data of the heating network are not within the abnormal range of their respective items.

[0042] In one embodiment of this application, after obtaining the analysis results of a heating network with potential for failure, the neural network-driven heating network failure analysis method further includes: The control parameter combinations for the heating network under no-fault risk conditions and the control parameter combinations for the heat source under normal operation are obtained, and then combined to obtain the normal control parameter combination for the heating network. In one embodiment of this application, the method for determining whether the heat source is operating normally can be the same as the method for determining whether the heating network has a fault risk, or it can be determined by monitoring real-time operating status data; no specific limitation is made here. Both the control parameter combinations for the heating network under no-fault risk conditions and the control parameter combinations for the heat source under normal operation can be determined and recorded according to a predetermined monitoring cycle.

[0043] Based on the normal control parameter combination of the heating network, the heat generated by the heating network within a preset time is determined and recorded as the first heat; and the production cost of the first heat is determined based on the first heat. In one embodiment of this application, before determining the heat generated by the heating network within a preset time based on the normal control parameter combination of the heating network, the normal control parameters of the heating network during historical operation and the heat generated by the heating network within the preset time are recorded; and the normal control parameters of the heating network during historical operation and the heat generated by the heating network within the preset time are fitted to obtain the heat curve generated by the heating network under the control of the normal control parameters over the preset time. Thus, after obtaining the normal control parameter combination and the preset time, the first heat can be determined. After obtaining the first heat, the first heat is multiplied by the cost required for the heating network to generate a unit of heat per unit time to obtain the first cost of the first heat. The cost required for the heating network to generate a unit of heat per unit time is estimated based on fuel costs, heat conversion efficiency of heat sources, pipeline length, heat loss rate of pipelines, heat exchange efficiency of heat exchange stations, heat dissipation efficiency of radiators, equipment maintenance costs, etc., or can be set based on empirical data. A centralized heating system includes: heat source, heating network, radiators, etc.

[0044] Based on the first heat capacity and the production cost of the first heat capacity, evaluation parameters for the normal control parameter combination of the heating network are determined. In one embodiment of this application, the evaluation parameters for the normal control parameter combination of the heating network include: Equation (5) in, Evaluation parameters representing the normal control parameter combination of the heating network. This indicates the heating capacity of the centralized heating system, determined by the heat generated within a preset time period. This represents the adjustment coefficient. The heat cost rate of a centralized heating system is the ratio of the production cost of the heat generated within a preset time period to the heat generated within the preset time period.

[0045] Based on the evaluation parameters of the normal control parameter combinations, the normal control parameter combinations are screened to obtain the first screened control parameter combinations. In one embodiment of this application, the evaluation parameters of the normal control parameter combinations are sorted, and normal control parameter combinations with a preset number of evaluation parameters are selected from the normal control parameter combinations as the first screened control parameter combinations.

[0046] A target control parameter combination is determined based on the evaluation parameters of the first screening control parameter combination and the differences between the control data in the first screening control parameter combination and the control data in the control data combination. In one embodiment of this application, the process of determining the target control parameter combination based on the evaluation parameters of the first screening control parameter combination and the differences between the control data in the first screening control parameter combination and the control data in the control data combination includes: recording the differences between the control data in the first screening control parameter combination and the control data in the control data combination as a first difference; if the first difference is greater than or equal to a preset difference threshold, updating the first screening control parameter combination and determining the target control parameter combination based on the updated control parameter combination; if the first difference is less than the preset difference threshold, using the first screening control parameter combination with the calculated first difference as a second screening control parameter combination; if the number of second screening control parameter combinations is greater than a preset number threshold, using the second screening control parameter combination with the largest evaluation parameter as the target control parameter combination; if the number of second screening control parameter combinations is less than or equal to the preset number threshold, using the second screening control parameter combination as the target control parameter combination.

[0047] In one embodiment of this application, after determining that there is a risk of failure in the operation of the heating network, the risk of failure is reduced and the safety of the heating network operation is improved by changing the combination of control parameters of the heating network.

[0048] In one embodiment of this application, the process of determining the target control parameter combination based on the evaluation parameters of the first screening control parameter combination and the differences between the control data in the first screening control parameter combination and the control data in the control data combination includes: The difference between the control data in the first selection control parameter combination and the control data in the control data combination is denoted as the first difference. In one embodiment of this application, the formula for calculating the first difference includes: Equation (6) in, Indicates the first difference, Indicates the first Weighting coefficients for control data items Indicates the first in the first combination of screening control parameters Item control data, Indicates the first in the control data combination Item control data. If the control data in the first screening control parameter combination is a data range, the first difference calculation is the difference between the control data in the control data combination and the upper or lower limit of the corresponding item data range in the first screening control parameter combination. Specifically, if the difference between the control data in the control data combination and the upper limit of the corresponding item data range in the first screening control parameter combination is the smallest, then the difference between the control data in the control data combination and the upper limit of the corresponding item data range in the first screening control parameter combination is taken as the first difference; if the difference between the control data in the control data combination and the lower limit of the corresponding item data range in the first screening control parameter combination is the smallest, then the difference between the control data in the control data combination and the lower limit of the corresponding item data range in the first screening control parameter combination is taken as the first difference.

[0049] If the first difference is greater than or equal to a preset difference threshold, the first screening control parameter combination is updated, and the target control parameter combination is determined based on the updated control parameter combination. In one embodiment of this application, the preset difference threshold is set according to the actual situation. The process of updating the first screening control parameter combination and determining the target control parameter combination based on the updated control parameter combination includes: taking the first screening control parameter combination as a population individual, and taking the evaluation parameter of the first screening control parameter combination as the fitness of the population individual; performing crossover and mutation on the population individuals to obtain updated population individuals; and screening the updated population individuals based on their fitness to obtain screened population individuals; and determining the target control parameter combination based on the fitness of the screened population individuals and the difference between the control data in the screened population individuals and the control data in the control data combination.

[0050] If the first difference is less than a preset difference threshold, then the first selection control parameter combination with the calculated first difference is used as the second selection control parameter combination. In one embodiment of this application, if the first difference is less than the preset difference threshold, it indicates that the difference between the control data in the first selection control parameter combination and the control data in the control data combination is small. When switching the control data in the first selection control parameter combination and the control data in the control data combination, the fluctuation of equipment operating status or equipment load in the centralized heating system can be effectively reduced.

[0051] If the number of second screening control parameter combinations exceeds a preset threshold, the second screening control parameter combination with the largest evaluation parameter is selected as the target control parameter combination. In one embodiment of this application, the preset threshold is set according to actual conditions. For example, setting the preset threshold to 1 and selecting the second screening control parameter combination with the largest evaluation parameter as the target control parameter combination can not only reduce fluctuations in the operating status or load of equipment in the centralized heating system, but also take into account the heat cost rate and heating efficiency of the centralized heating system, which is beneficial for saving heating costs and improving heating efficiency.

[0052] If the number of second screening control parameter combinations is less than or equal to a preset threshold, then the second screening control parameter combination is used as the target control parameter combination. In one embodiment of this application, using the second screening control parameter combination as the target control parameter combination when the number of second screening control parameter combinations is less than or equal to a preset threshold is beneficial for reducing fluctuations in the operating status or load of equipment in a centralized heating system.

[0053] In one embodiment of this application, the process of updating the first screening control parameter combination and determining the target control parameter combination based on the updated control parameter combination includes: The first selection control parameter combination is used as the individual population, and the evaluation parameter of the first selection control parameter combination is used as the fitness of the individual population. In one embodiment of this application, if the number of the first selection control parameter combinations is one set, the first selection control parameter combinations are expanded by adding preset control parameter combinations; if the number of the first selection control parameter combinations is multiple sets, it is not necessary to expand the first selection control parameter combinations, thereby always maintaining the diversity of the first selection control parameter combinations.

[0054] Crossover and mutation are performed on individuals in the population to obtain updated population individuals; and the updated population individuals are then screened based on their fitness to obtain screened population individuals. In one embodiment of this application, the process of crossover of population individuals includes: randomly selecting one or more pairs of parent population individuals from the updated population individuals according to a preset ratio; selecting a crossover point and exchanging genes in the parent population individuals according to the crossover method to obtain crossovered population individuals; and mutating gene segments in the crossovered population individuals to obtain updated population individuals. The format of the parent population individuals is as follows: and ,in, This indicates the name of the current control parameter 1 for the heat source in the first generation population. Indicates the name of the current control parameter 2 for the heat source in the first generation population. This indicates the name of the current control parameter 3 for the heat source in the first generation population. This indicates the name of the current control parameter 1 for the heat network in the first generation population. This indicates the name of the current control parameter 2 for the heat network in the first generation of the population. This indicates the name of the current control parameter 3 for the heat network in the first generation of the population. This indicates the current control parameter name 1 for the heat source in the second generation population. This indicates the name of the current control parameter 2 for the heat source in the second generation population. Indicates the name of the current control parameter 3 for the heat source in the second generation population. This indicates the name of the current control parameter 1 for the heat network in the second generation population. This indicates the name of the current control parameter 2 for the heat network in the second generation population. Let 3 represent the current control parameter name of the heat network in the second parent population. Each parameter name in the parent population contains parameter data. The crossover point is set to the position between the current control parameter name 3 of the heat source and the current control parameter name 1 of the heat network. The crossover method is set to single-point crossover. The resulting population is then... and In the crossover population, the types of control parameters remain unchanged compared to the parent population; however, the data values ​​or ranges of these control parameters change. The process of mutating gene segments in the crossover population to obtain updated population individuals involves: using the data of each control parameter in the updated population individual as a gene segment, and altering the data of each control parameter in the updated population individual according to a preset mutation probability to obtain the updated population individual. The process of modifying the data of each control parameter in the updated population includes: setting a preset change amount for the control parameter data; if the control parameter data is in the form of a data value, then the sum of the control parameter data and the preset change amount is used as the mutated control parameter data, or the difference between the control parameter data and the preset change amount is used as the mutated control parameter data; if the control parameter data is in the form of a data range, then the difference between the upper limit of the control parameter data range and the preset change amount, or the sum of the upper limit of the control parameter data range and the preset change amount, is used as the upper limit of the mutated control parameter data range; and the difference between the lower limit of the control parameter data range and the preset change amount, or the sum of the lower limit of the control parameter data range and the preset change amount, is used as the lower limit of the mutated control parameter data range.

[0055] In one embodiment of this application, after obtaining the updated population individuals, the data of each control parameter in the updated population individuals needs to be compared with the control parameter range of the corresponding item in the combination of abnormal control parameter ranges for the heating network. If the data of each control parameter in the updated population individual falls within the control parameter range of the corresponding item in the combination of abnormal control parameter ranges for the heating network, then the population individual is discarded. This setting ensures that the updated population individuals do not belong to the abnormal control parameter reference range combination, which is beneficial for ensuring the safety of the heating network during operation.

[0056] In one embodiment of this application, the process of screening individuals in the updated population based on their fitness to obtain the screened individuals is the same as the process of screening the normal control parameter combination based on the evaluation parameters of the normal control parameter combination to obtain the first screening control parameter combination.

[0057] The target control parameter combination is determined based on the fitness of individuals in the screened population and the differences between the control data in the screened population and the control data in the control data combination. In one embodiment of this application, the process of determining the target control parameter combination based on the fitness of individuals in the screened population and the differences between the control data in the screened population and the control data in the control data combination is the same as the process of determining the target control parameter combination based on the evaluation parameters of the first screening control parameter combination and the differences between the control data in the first screening control parameter combination and the control data in the control data combination. The process of determining the target control parameter combination based on the fitness of individuals in the screened population and the differences between the control data in the screened population and the control data in the control data combination can be repeated until the target control parameter combination is found.

[0058] In one embodiment of this application, before acquiring the current operating status data and current control parameter changes of the heat source, and the current operating status data, current control parameter changes, and fault determination conditions of the heating network, the neural network-driven heating network fault analysis method further includes: The system collects the current heating capacity, current outlet water supply temperature, current outlet water supply pressure, current outlet water supply flow rate, current inlet return water temperature, current fuel consumption control value, current equipment load rate control value, and current supply and return water temperature difference control value during the operation of the heat source; and collects the current outlet water supply temperature, current outlet water supply pressure, current outlet water supply flow rate, current inlet return water temperature, current circulating pump operating frequency control value, current flow valve opening control value, and current supply and return water temperature difference control value during the operation of the heating network. In one embodiment of this application, the current outlet water supply temperature during the operation of the heat source is collected by a temperature sensor, the current outlet water supply pressure during the operation of the heat source is collected by a pressure sensor, the current outlet water supply flow rate during the operation of the heat source is collected by a flow sensor, and the current inlet return water temperature during the operation of the heat source is collected by a temperature sensor. The current heating capacity, current fuel consumption control value, and current equipment load rate control value during the operation of the heat source are obtained by pre-setting. The current outlet water supply temperature during the operation of the heating network is obtained by a temperature sensor; the current outlet water supply pressure during the operation of the heating network is obtained by a pressure sensor; the current outlet water supply flow rate during the operation of the heating network is obtained by a flow sensor; the current inlet return water temperature during the operation of the heating network is obtained by a temperature sensor; and the current circulating pump operating frequency control value, the current flow valve opening control value, and the current supply and return water temperature difference control value during the operation of the heating network are obtained by pre-setting.

[0059] The current heat supply, current outlet water supply temperature, current outlet water supply pressure, current outlet water supply flow rate, and current inlet return water temperature during the operation of the heat source are used as the current operating status data of the heat source; the current fuel consumption control value, current equipment load rate control value, and current supply and return water temperature difference control value during the operation of the heat source are used as the current control parameter data of the heat source. In one embodiment of this application, the current operating status data of the heat source is used to reflect the operating status of the heat source under the control of the current control parameter data of the heat source.

[0060] The current outlet water supply temperature, current outlet water supply pressure, current outlet water supply flow rate, and current inlet return water temperature during the operation of the heating network are used as the current operating status data of the heating network; the current circulating pump operating frequency control value, current flow valve opening control value, and current supply and return water temperature difference control value during the operation of the heating network are used as the current control parameter data of the heating network. In one embodiment of this application, the current operating status data of the heating network is used to reflect the operating status of the heating network under the control of the current control parameter data of the heating network and the current control parameter data of the heat source.

[0061] When a heat source or heating network malfunctions, the current control parameter data of the heat source and the current control parameter data of the heating network are combined to obtain a combined range of abnormal control parameters for the heating network; and the abnormal range of the current operating status data of the heating network is used as the abnormal status data range of the heating network. In one embodiment of this application, the abnormal range of the current operating status data of the heating network can be obtained by pre-setting.

[0062] Figure 3 This is a schematic diagram illustrating the structure of a heating network operation status output model, as shown in an exemplary embodiment of this application. Figure 3 As shown, the output model of the heating network operation status includes an input layer, a residual module, a pooling layer, and a fully connected layer. The input layer is used to determine the changes in the historical operation status data of the heat source and the heating network. The residual module extracts data association features based on the changes in the historical operation status data of the heat source, the changes in the historical control parameters of the heat source, and the changes in the historical operation status data and the historical control parameters of the heating network. The pooling layer performs dimensionality reduction, compression, and noise smoothing operations on the data association features using aggregation functions to obtain the reduced-dimensional data association features. The fully connected layer integrates the reduced-dimensional data association features to obtain the data association function.

[0063] The following describes an embodiment of the apparatus described in this application, which can be used to execute the neural network-driven heating network fault analysis method described in the above embodiments of this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the neural network-driven heating network fault analysis method described above in this application.

[0064] Figure 4 This is a block diagram illustrating a neural network-driven thermal network fault analysis system, as shown in an exemplary embodiment of this application.

[0065] like Figure 4 As shown, the exemplary neural network-driven heating network fault analysis system 400 includes: The data acquisition module 401 is used to acquire the current operating status data and current control parameter changes of the heat source, as well as the current operating status data, current control parameter changes, and fault judgment conditions of the heating network.

[0066] The data combination module 402 is used to combine the current control parameter data of the heat source and the current control parameter data of the heating network to obtain the control data combination.

[0067] The change determination module 403 is used to input the current operating status data of the heat source, the change in the current control parameters of the heat source, and the change in the current control parameters of the heat network into the heat network operating status output model if the control data combination does not match the range combination of abnormal control parameters of the heat network, and the current operating status data of the heat network does not match the range of abnormal status data of the heat network, so as to obtain the change in the operating status data of the heat network.

[0068] The fault analysis module 404 is used to take the current operating status data of the heating network and the sum of the changes in the operating status data of the heating network as the updated status data of the heating network; and to determine the fault analysis result of the heating network based on the matching result of the updated status data of the heating network and the range of abnormal status data of the heating network.

[0069] In one embodiment of this application, the current operating status data of the heat source includes: current heat supply, current outlet water supply temperature, current outlet water supply pressure, current outlet water supply flow rate, and current inlet return water temperature during the heat source operation process. The current control parameter data of the heat source includes: current fuel consumption control value, current equipment load rate control value, and current supply and return water temperature difference control value during the heat source operation process. The current operating status data of the heating network includes: current outlet water supply temperature, current outlet water supply pressure, current outlet water supply flow rate, and current inlet return water temperature during the heating network operation process. The current control parameter data of the heating network includes: current circulating pump operating frequency control value, current flow valve opening control value, and current supply and return water temperature difference control value during the heating network operation process. Fault determination conditions include: a combination of abnormal heating network status data ranges and abnormal heating network control parameter ranges. The abnormal heating network status data ranges include: abnormal ranges for current heat supply, current outlet water supply temperature, current outlet water supply pressure, current outlet water supply flow rate, and current inlet return water temperature. The range of abnormal status data for the heating network needs to be set according to the design requirements of the heating network. The combination of abnormal control parameter ranges for the heating network includes: abnormal control ranges for fuel consumption of the heat source, abnormal control ranges for equipment load rate of the heat source, abnormal control ranges for supply and return water temperature differences of the heat source, abnormal control ranges for the operating frequency of the circulating pumps in the heating network, abnormal control ranges for the opening degree of the flow valves in the heating network, and abnormal control ranges for supply and return water temperature differences in the heating network. The abnormal control range for fuel consumption of the heat source consists of the fuel consumption when the heat source or the heating network experiences a fault; the abnormal control range for equipment load rate of the heat source consists of... The equipment load rate is composed of the following parameters when the heat source or heating network fails: the supply and return water temperature difference control range is composed of the supply and return water temperature difference when the heat source or heating network fails; the circulating pump operating frequency control range is composed of the circulating pump operating frequency when the heat source or heating network fails; the flow valve opening degree control range is composed of the flow valve opening degree when the heat source or heating network fails; and the supply and return water temperature difference control range is composed of the supply and return water temperature difference when the heat source or heating network fails.The change in the current control parameters of the heat source is determined by the control parameter data of the heat source at the previous moment and the current control parameter data. That is, the change in the current control parameters of the heat source is the difference between the current control parameter data and the control parameter data of the heat source at the previous moment. If the control parameter data of the heat source includes multiple items, the sum of the absolute values ​​of the differences between the current and previous moments for each control parameter is taken as the change in the current control parameters of the heat source. The change in the current control parameters of the heating network is determined by the control parameter data of the heating network at the previous moment and the current control parameter data. That is, the change in the current control parameters of the heating network is the difference between the current and previous moments for the heating network. If the control parameter data of the heating network includes multiple items, the sum of the absolute values ​​of the differences between the current and previous moments for each control parameter is taken as the change in the current control parameters of the heating network.

[0070] In one embodiment of this application, the process of combining the current control parameter data of the heat source and the current control parameter data of the heating network is as follows: the current control parameter data of the heat source and the current control parameter data of the heating network are combined sequentially, and the format of the control data combination is [current control parameter name 1 of heat source: data of current control parameter name 1 of heat source; current control parameter name 2 of heat source: data of current control parameter name 2 of heat source; current control parameter name 3 of heat source: data of current control parameter name 3 of heat source; current control parameter name 1 of heating network: data of current control parameter name 1 of heating network...].

[0071] In one embodiment of this application, the heating network operation status output model is obtained by training a pre-constructed operation status output model based on sample data. The process of matching control data combinations with the range of abnormal control parameters in the heating network includes: comparing each control parameter in the control data combination with the corresponding abnormal control parameter range in the range of abnormal control parameters in the heating network; if each control parameter in the control data combination falls within the corresponding abnormal control parameter range in the range of abnormal control parameters in the range of abnormal control parameters in the heating network, then the control data combination is determined to match the range of abnormal control parameters in the heating network; if at least one control parameter in the control data combination does not fall within the corresponding abnormal control parameter range in the range of abnormal control parameters in the range of abnormal control parameters in the heating network, then the control data combination is determined to not match the range of abnormal control parameters in the heating network. The process of matching the current operating status data of the heating network with the range of abnormal status data in the heating network includes: comparing the current operating status data of the heating network with the range of abnormal status data in the heating network; if the current operating status data of the heating network falls within the range of abnormal status data in the heating network, then the current operating status data of the heating network is determined to match the range of abnormal status data in the heating network; if the current operating status data of the heating network does not fall within the range of abnormal status data in the heating network, then the current operating status data of the heating network is determined to not match the range of abnormal status data in the heating network. If the current operating status data of the heating network contains multiple types of operating status data, then each type of operating status data is compared with the corresponding type of abnormal status data range in the heating network abnormal status data range. If any type of operating status data falls within the corresponding type of abnormal status data range, then the current operating status data of the heating network is determined to match the heating network abnormal status data range. If none of the operating status data types fall within the corresponding type of abnormal status data range, then the current operating status data of the heating network is determined to not match the heating network abnormal status data range. If the control data combination does not match the heating network abnormal control parameter range combination, and the current operating status data of the heating network does not match the heating network abnormal status data range, then the operating status of the heating network is considered normal.

[0072] In one embodiment of this application, the determination that the control data combination does not match the range of abnormal control parameters of the heating network, and that the current operating status data of the heating network does not match the range of abnormal status data of the heating network, is performed under the condition that the heat source and the radiator are operating normally. Whether the operating status of the heating network is normal is determined by whether the operating status data of the heating network is within the normal range, and whether the operating status of the radiator is normal is determined by whether the operating status data of the radiator is within the normal range.

[0073] In one embodiment of this application, the normal operation of the heating network is determined by judging whether the combination of control data matches the combination of abnormal control parameters of the heating network. This avoids the lag caused by comparing real-time operating status data with the range of abnormal operating status data, and reduces the probability of abnormal operation of the heating network caused by potential fault risks or dangerous factors.

[0074] In one embodiment of this application, when the current control parameters of the heat source change, and / or the current control parameters of the heating network change, and under the condition that the operating status of the heating network and the operating status of the heat source are both normal, the change in the current operating status data of the heating network caused by the change in the current control parameters of the heat source and / or the change in the current control parameters of the heating network is promptly known. Thus, based on the matching result of the updated status data of the heating network and the range of abnormal status data of the heating network, the heating network fault analysis result is determined in a timely manner, and the occurrence of faults caused by potential fault risks or dangerous factors is avoided.

[0075] It should be noted that the neural network-driven heating network fault analysis system and the neural network-driven heating network fault analysis method provided in the above embodiments belong to the same concept. The specific operation methods of each module and unit have been described in detail in the method embodiments and will not be repeated here. In practical applications, the neural network-driven heating network fault analysis system provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. This is not a limitation here.

[0076] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

Claims

1. A neural network-driven method for analyzing heating network faults, characterized in that, The method includes: The system acquires the current operating status data and current control parameter changes of the heat source, the current operating status data, current control parameter changes, and fault determination conditions of the heating network. The fault determination conditions include: a range of abnormal heating network status data and a combination of abnormal heating network control parameter ranges. The current control parameter changes of the heat source are determined by the control parameter data of the heat source at the previous moment and the current control parameter data. The current control parameter changes of the heating network are determined by the control parameter data of the heating network at the previous moment and the current control parameter data. The current control parameter data of the heat source and the current control parameter data of the heating network are combined to obtain the control data combination; If the control data combination does not match the range of abnormal control parameters of the heating network, and the current operating status data of the heating network does not match the range of abnormal status data of the heating network, then the current operating status data of the heat source, the change in the current control parameters of the heat source, and the change in the current control parameters of the heating network are input into the heating network operating status output model to obtain the change in the operating status data of the heating network; the heating network operating status output model is obtained by training a pre-constructed operating status output model based on sample data; The sum of the current operating status data of the heating network and the change in the operating status data of the heating network is used as the updated status data of the heating network; the heating network fault analysis result is determined based on the matching result between the updated status data of the heating network and the range of abnormal status data of the heating network.

2. The neural network-driven heating network fault analysis method according to claim 1, characterized in that, If the pre-built operating status output model includes: an input layer, a residual module, a pooling layer, and a fully connected layer, and the sample data includes: historical operating status data and historical control parameter changes of the heat source, and historical operating status data and historical control parameter changes of the heating network, then the process of training the pre-built operating status output model based on the sample data to obtain the heating network operating status output model includes: The historical operating status data of the heat source and the historical operating status data of the heating network are input into the input layer to obtain the change amount of the historical operating status data of the heat source and the change amount of the historical operating status data of the heating network. The changes in the historical operating status data of the heat source, the changes in the historical control parameters of the heat source, the changes in the historical operating status data of the heating network, and the changes in the historical control parameters of the heating network are input into the residual module to obtain data association features. The data association features include: the association function between the control parameters of the heat source and the operating status data of the heat source, the association function between the operating status data of the heat source and the operating status data of the heating network, and the association function between the control parameters of the heating network and the operating status data of the heating network. The data association features are input into the pooling layer to obtain data association features with reduced dimensionality. The reduced-dimensional data association features are input into the fully connected layer to obtain the data association function; the data association function includes: the association function between the control parameters of the heat source, the operating status data of the heat source, the operating status data of the heating network, and the control parameters of the heating network; By verifying the data and the data association function, the parameters in the pre-built operational transition output model are adjusted to obtain the heating network operational status output model.

3. The neural network-driven heating network fault analysis method according to claim 1 or 2, characterized in that, The process of determining the results of the heating network fault analysis based on the matching results between the updated status data of the heating network and the range of abnormal status data of the heating network includes: If the updated status data of the heating network is within the range of the abnormal status data of the heating network, then it is determined that the updated status data of the heating network matches the range of the abnormal status data of the heating network, and the existence of fault risk is taken as the fault analysis result of the heating network. If the updated status data of the heating network is not within the range of the abnormal status data of the heating network, it is determined that the updated status data of the heating network does not match the range of the abnormal status data of the heating network, and the absence of fault risk is taken as the result of the fault analysis of the heating network.

4. The neural network-driven heating network fault analysis method according to claim 3, characterized in that, After obtaining the fault analysis results of the heating network with potential failure risks, the method further includes: Obtain the control parameter combination of the heating network under the condition of no fault risk and the control parameter combination of the heat source under normal operation, and merge the control parameter combination of the heating network under the condition of no fault risk and the control parameter combination of the heat source under normal operation to obtain the normal control parameter combination of the heating network. Based on the normal control parameter combination of the heating network, the heat generated by the heating network within a preset time is determined and recorded as the first heat; and based on the first heat, the production cost of the first heat is determined. Based on the first heat volume and the production cost of the first heat volume, the evaluation parameters for the normal control parameter combination of the heating network are determined. Based on the evaluation parameters of the normal control parameter combination, the normal control parameter combination is screened to obtain a first screened control parameter combination; The target control parameter combination is determined based on the evaluation parameters of the first screening control parameter combination and the differences between the control data in the first screening control parameter combination and the control data in the control data combination.

5. The neural network-driven heating network fault analysis method according to claim 4, characterized in that, The evaluation parameters for the normal control parameter combination of the heating network include: ， in, Evaluation parameters representing the normal control parameter combination of the heating network. This indicates the heating capacity of the centralized heating system, determined by the heat generated within a preset time period. This represents the adjustment coefficient. The heat cost rate of a centralized heating system is determined by the heat generated within a preset time period and the production cost of the heat generated within the preset time period.

6. The neural network-driven heating network fault analysis method according to claim 4, characterized in that, The process of determining the target control parameter combination based on the evaluation parameters of the first screening control parameter combination and the differences between the control data in the first screening control parameter combination and the control data in the control data combination includes: The difference between the control data in the first combination of screening control parameters and the control data in the control data combination is denoted as the first difference; If the first difference is greater than or equal to a preset difference threshold, the first screening control parameter combination is updated, and the target control parameter combination is determined based on the updated control parameter combination. If the first difference is less than the preset difference threshold, then the first combination of screening control parameters calculated from the first difference is used as the second combination of screening control parameters. If the number of the second screening control parameter combinations is greater than the preset number threshold, then the second screening control parameter combination with the largest evaluation parameter will be used as the target control parameter combination. If the number of the second filtering control parameter combinations is less than or equal to the preset number threshold, then the second filtering control parameter combination is used as the target control parameter combination.

7. The neural network-driven heating network fault analysis method according to claim 6, characterized in that, The process of updating the first selection control parameter combination and determining the target control parameter combination based on the updated control parameter combination includes: The first combination of screening control parameters is used as the population individual, and the evaluation parameter of the first combination of screening control parameters is used as the fitness of the population individual. Crossover and mutation are performed on the individuals in the population to obtain an updated population; and the updated population individuals are screened based on their fitness to obtain a screened population. The target control parameter combination is determined based on the fitness of the individuals in the screened population and the difference between the control data in the individuals in the screened population and the control data in the control data combination.

8. The neural network-driven heating network fault analysis method according to claim 6, characterized in that, The formula for calculating the first difference includes: ， in, Indicates the first difference, Indicates the first Weighting coefficients for control data items Indicates the first in the first combination of screening control parameters Item control data, Indicates the first in the control data combination Item control data.

9. The neural network-driven heating network fault analysis method according to claim 1 or 2, characterized in that, Before acquiring the current operating status data and current control parameter changes of the heat source, and the current operating status data, current control parameter changes, and fault determination conditions of the heating network, the method further includes: The system collects the current heat supply, current outlet water supply temperature, current outlet water supply pressure, current outlet water supply flow rate, current inlet return water temperature, current fuel consumption control value, current equipment load rate control value, and current supply and return water temperature difference control value during the operation of the heat source; and collects the current outlet water supply temperature, current outlet water supply pressure, current outlet water supply flow rate, current inlet return water temperature, current circulating pump operating frequency control value, current flow valve opening control value, and current supply and return water temperature difference control value during the operation of the heating network. The current heat supply, current outlet water supply temperature, current outlet water supply pressure, current outlet water supply flow rate, and current inlet return water temperature during the operation of the heat source are used as the current operating status data of the heat source; the current fuel consumption control value, current equipment load rate control value, and current supply and return water temperature difference control value during the operation of the heat source are used as the current control parameter data of the heat source. The current outlet water supply temperature, current outlet water supply pressure, current outlet water supply flow rate, and current inlet return water temperature during the operation of the heating network are used as the current operating status data of the heating network; the current circulating pump operating frequency control value, current flow valve opening control value, and current supply and return water temperature difference control value during the operation of the heating network are used as the current control parameter data of the heating network. When a fault occurs in the heat source or the heating network, the current control parameter data of the heat source and the current control parameter data of the heating network are combined to obtain the abnormal control parameter range combination of the heating network; and the abnormal range of the current operating status data of the heating network is taken as the abnormal status data range of the heating network.

10. A neural network-driven thermal network fault analysis system, characterized in that, include: The data acquisition module is used to acquire the current operating status data and current control parameter changes of the heat source, as well as the current operating status data, current control parameter changes, and fault judgment conditions of the heating network. The fault determination conditions include: a combination of abnormal state data ranges and abnormal control parameter ranges of the heating network; the change in the current control parameter of the heat source is determined by the control parameter data of the heat source at the previous moment and the current control parameter data; the change in the current control parameter of the heating network is determined by the control parameter data of the heating network at the previous moment and the current control parameter data. The data combination module is used to combine the current control parameter data of the heat source and the current control parameter data of the heating network to obtain control data combination; The change determination module is used to input the current operating status data of the heat source, the change in the current control parameters of the heat source, and the change in the current control parameters of the heat network into the heat network operating status output model if the control data combination does not match the range combination of abnormal control parameters of the heating network, and the current operating status data of the heating network does not match the range of abnormal state data of the heating network, so as to obtain the change in the operating status data of the heating network; the heat network operating status output model is obtained by training a pre-constructed operating status output model based on sample data; The fault analysis module is used to take the current operating status data of the heating network and the sum of the changes in the operating status data of the heating network as the updated status data of the heating network; and to determine the fault analysis result of the heating network based on the matching result between the updated status data of the heating network and the range of abnormal status data of the heating network.

Images