Method and device for automatic generation of energy system heat balance diagram

Abstract

The application discloses an automatic generation method and device of an energy system heat balance diagram, and belongs to the technical field of energy systems. The method comprises the following steps: constructing a basic database based on historical project data of a target energy system, and obtaining a plurality of training samples; the training samples comprise data of boundary parameters and characteristic parameters as input and a heat balance diagram as output; the heat balance model is trained by using the plurality of training samples, so that the heat balance model determines a graph ontology based on the data of the boundary parameters in the training samples, predicts the thermal parameters of each graph node in the graph ontology based on the data of the boundary parameters and the data of the characteristic parameters in the training samples, adjusts the model parameters based on the loss between the prediction results and the thermal parameters of the graph nodes on the heat balance diagram in the training samples, and obtains the trained heat balance model; and the heat balance model is used to output a heat balance diagram for input data of a heat balance diagram to be generated. The application can automatically obtain a required heat balance diagram.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of energy systems, and in particular to an automatic generation method and device for a heat balance diagram of an energy system. BACKGROUND

[0002] Heat balance diagram calculation is a core link in thermodynamic analysis and energy system design, and is widely used in process calculation of complex energy systems such as thermal power generation, gas-steam combined cycle, nuclear power generation, compressed air energy storage system, etc. The heat balance diagram describes the flow and transformation relationship of energy and matter in the system, and provides an important theoretical basis for system design, optimization and operation. The calculation results can provide the thermodynamic parameters of each node in the process, including pressure, temperature, enthalpy, flow rate and other key data. These data are the "command diagram" for subsequent system detail design, and directly determine the performance, efficiency and reliability of the system. Therefore, the accuracy and speed of heat balance diagram calculation have a crucial impact on the execution of engineering projects.

[0003] In related technologies, heat balance diagram calculation needs to rely heavily on the experience and knowledge of designers, which is not only inefficient, but also prone to human error. SUMMARY

[0004] The present application provides an automatic generation method and device for a heat balance diagram of an energy system. The technical solution is as follows:

[0005] On the one hand, an automatic generation method for a heat balance diagram of an energy system is provided, which comprises:

[0006] Based on the historical project data of the target energy system, a basic database is constructed; the basic database includes a plurality of types of heat balance diagrams corresponding to the target energy system; the heat balance diagram includes a graph ontology and thermodynamic parameters labeled on the graph nodes, and the graph ontology is formed by a plurality of graph nodes and their connection relationships;

[0007] Based on the basic database, a plurality of training samples are obtained; the training samples include data of boundary parameters and characteristic parameters as input, and also include heat balance diagrams as output;

[0008] A heat balance model is trained using a plurality of training samples, so that the heat balance model determines the graph ontology of the heat balance diagram to be output based on the data of the boundary parameters in the training samples, and predicts the thermodynamic parameters of each graph node in the graph ontology based on the data of the boundary parameters and the data of the characteristic parameters in the training samples. The model parameters are adjusted based on the loss between the prediction results and the thermodynamic parameters of the graph nodes in the heat balance diagram in the training samples, and a trained heat balance model is obtained;

[0009] The heat balance model is used to output a heat balance diagram for input data of the heat balance diagram to be generated; the input data is data input according to boundary parameters and characteristic parameters.

[0010] In another aspect, an automatic generation device of a heat balance diagram of an energy system is provided, and the device comprises:

[0011] A construction unit is configured to construct a basic database based on historical project data of a target energy system; the basic database comprises a plurality of types of heat balance diagrams corresponding to the target energy system; the heat balance diagram comprises a graph ontology and thermal parameters labeled on graph nodes, and the graph ontology is formed by a plurality of graph nodes and connection relationships therebetween;

[0012] An acquisition unit is configured to acquire a plurality of training samples based on the basic database; the training sample comprises data of boundary parameters and characteristic parameters as input, and further comprises a heat balance diagram as output;

[0013] A training unit is configured to train a heat balance model by using the plurality of training samples, so that the heat balance model determines a graph ontology of a heat balance diagram to be output based on data of boundary parameters in the training sample, and predicts thermal parameters of each graph node in the graph ontology based on data of boundary parameters and data of characteristic parameters in the training sample, adjusts model parameters based on a loss between a prediction result and thermal parameters of graph nodes in the heat balance diagram in the training sample, and obtains a trained heat balance model;

[0014] A heat balance calculation unit is configured to output a heat balance diagram for input data of the heat balance diagram to be generated by using the heat balance model; the input data is data input according to boundary parameters and characteristic parameters.

[0015] In another aspect, a computer device is provided, and the computer device comprises a memory and a processor; the memory is configured to store a computer program; and the processor is configured to execute the computer program stored in the memory, so as to implement steps of the automatic generation method of the heat balance diagram of the energy system.

[0016] In another aspect, a computer readable storage medium is provided, and the storage medium stores a computer program; when the computer program is executed by a processor, steps of the automatic generation method of the heat balance diagram of the energy system are implemented.

[0017] In another aspect, a computer program product is provided, and the computer program product comprises a computer program; when the computer program is executed by a processor, steps of the automatic generation method of the heat balance diagram of the energy system are implemented.

[0018] The technical solution provided by the present application can at least bring the following beneficial effects:

[0019] By constructing a basic database by using historical project information of a target energy system, and then obtaining training samples from the basic database, the heat balance model is trained, and after the training is completed, only the data of boundary parameters and the data of characteristic parameters are input into the trained heat balance model, so that the heat balance diagram output by the heat balance model can be obtained. According to the scheme, the operator can adjust and input the data of the input boundary parameters and characteristic parameters, so that the required heat balance diagram can be obtained automatically and quickly, without the need for the operator to have design experience, and the accuracy and the generation efficiency can be ensured. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0021] Figure 1 is a flow chart of an automatic generation method of an energy system heat balance diagram provided by an embodiment of the present application;

[0022] Figure 2 is a structure diagram of a heat balance model provided by an embodiment of the present application;

[0023] Figure 3 is a structure diagram of an automatic generation device of an energy system heat balance diagram provided by an embodiment of the present application;

[0024] Figure 4 is a hardware architecture diagram of a computer device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0026] Please refer to Figure 1 The automatic generation method of an energy system heat balance diagram provided by the embodiments of the present application comprises:

[0027] Step 100, constructing a basic database based on historical project data of a target energy system; the basic database includes a plurality of types of heat balance diagrams corresponding to the target energy system; the heat balance diagram includes a graph ontology formed by a plurality of graph nodes and their connection relationships, and a thermal parameter marked on a graph node;

[0028] Step 102, obtaining a plurality of training samples based on the basic database; the training sample includes data of boundary parameters and characteristic parameters as input, and also includes a heat balance diagram as output;

[0029] Step 104, training the heat balance model by using the plurality of training samples, so that the heat balance model determines the graph ontology of the heat balance diagram to be output based on the data of the boundary parameters in the training sample, and predicts the thermal parameter of each graph node in the graph ontology based on the data of the boundary parameters and the data of the characteristic parameters in the training sample, adjusts the model parameters based on the loss between the prediction result and the thermal parameter of the graph node in the heat balance diagram in the training sample, and obtains the trained heat balance model;

[0030] Step 106, outputting a heat balance diagram for input data of a heat balance diagram to be generated by using the heat balance model; the input data is data input according to boundary parameters and characteristic parameters.

[0031] In the embodiment of the present application, the basic database is constructed by using the historical project information of the target energy system, and then the training sample is obtained from the basic database, the heat balance model is trained, and after the training is completed, only the data of the boundary parameters and the data of the characteristic parameters need to be input into the trained heat balance model, and the heat balance diagram output by the heat balance model can be obtained. According to the scheme, the operator can adjust and input the data of the input boundary parameters and characteristic parameters, so that the required heat balance diagram can be quickly obtained, the operator does not need to have design experience, and the accuracy and generation efficiency can be guaranteed.

[0032] The execution mode of each step is described below. Figure 1

[0033] First, steps 100 and 102 are described simultaneously.

[0034] The calculation of the heat balance diagram not only needs historical project data, but also needs to rely on the experience of designers and specific software, and not all designers who need energy system heat balance diagrams can obtain historical project data, therefore, a method for automatically generating a heat balance diagram is needed.

[0035] ​Considering that the mechanism model of the energy system can be used to calculate the thermodynamic parameters, however, the thermodynamic parameters calculated directly using the mechanism model are inconsistent with the actual project data. It can be seen that in actual project applications, there are other complex correlations between the components in the energy system, and therefore, the AI model can be used to learn the characteristics of the complex correlations to improve the accuracy of automatically generating the heat balance diagram.

[0036] In order to realize the automatic generation of the heat balance diagram, it is necessary to construct a basic database based on the historical project data of the target energy system by a professional based on thermodynamic professional knowledge, which is used to provide training samples for training the heat balance model, so as to automatically generate the heat balance diagram by using the heat balance model.

[0037] In the embodiment of the present application, the target energy system can be at least a thermal power generation system, a compressed air energy storage system, a gas-steam combined cycle system or a nuclear power generation system.

[0038] Specifically, the target energy system can be classified based on the historical project data of the target energy system. For example, the above target energy system can be classified as follows:

[0039] The thermal power generation system can be classified into subcritical 300MW unit, supercritical 300MW unit, ultra-supercritical 300MW unit, subcritical 600MW unit, supercritical 600MW unit, ultra-supercritical 600MW unit, subcritical 1000MW unit, supercritical 1000MW unit, ultra-supercritical 1000MW unit, etc.

[0040] The compressed air energy storage system can be classified into medium temperature 100MW level, high temperature 100MW level, medium temperature 200MW level, high temperature 200MW level, medium temperature 300MW level, high temperature 300MW level, etc.

[0041] The gas-steam combined cycle system can be classified into different types of different capacities and configurations;

[0042] The nuclear power generation system can be classified into different types of different capacity unit reactors and steam cycle systems.

[0043] For each type of the target energy system, boundary parameters and characteristic parameters can be determined based on historical project information of the corresponding type, and a corresponding heat balance diagram can be constructed using data of the boundary parameters and data of the characteristic parameters, so that a basic database can be obtained. The basic database includes heat balance diagrams of multiple types corresponding to the target energy system; the heat balance diagram includes a graph body and thermal parameters marked on graph nodes, the graph body is formed by multiple graph nodes and their connection relationships; the graph nodes are used to represent components in the target energy system, and the connection relationships are used to represent parameter transmission relationships; the parameter transmission relationships include energy transmission relationships, mass transmission relationships and kinetic energy transmission relationships.

[0044] The boundary parameters are external constraints defined when modeling or analyzing the energy system, and are used to describe the input / output limits of the energy system interacting with the outside world. For example, for a regulating valve, the boundary parameters can include inlet pressure, inlet specific enthalpy, inlet flow rate, etc.; for a compressor, the boundary parameters can include inlet temperature, inlet dryness, inlet specific enthalpy, specific entropy, volumetric flow rate, pressure ratio, etc.; for a turbine pressure unit, the boundary parameters can include inlet steam pressure, inlet steam specific enthalpy, unit flow rate, back pressure, etc.

[0045] The characteristic parameters are key variables that describe the inherent properties or dynamic behaviors of the energy system, and are used to reflect the physical, chemical or functional characteristics of the energy system itself. For example, for a regulating valve, the characteristic parameters can include relative pressure loss, low-pressure leakage area, low-pressure leakage flow coefficient, high-pressure leakage area, high-pressure leakage flow coefficient; for a compressor, the characteristic parameters can include isentropic efficiency, etc.; for a turbine pressure unit, the characteristic parameters can include isentropic efficiency, through-flow area, through-flow capacity correction coefficient, etc.

[0046] In the embodiment of the present application, when the boundary parameters are determined, the components included in the energy system and the connection relationships between the components can be determined, the connection relationships are parameter transmission relationships between the components in the energy system, and each component is used to represent a graph node, therefore, the boundary parameters determine the graph body in the heat balance diagram. In addition, the graph bodies of the heat balance diagrams of the same type of the target energy system are the same.

[0047] After the above basic database is constructed, training samples can be directly extracted. The training samples include data of the boundary parameters and data of the characteristic parameters as input, and further include the heat balance diagram as output.

[0048] Then, steps 104 and 106 are described simultaneously.

[0049] Since the boundary parameters determine the graph ontology of the heat balance diagram, in the embodiment of the present application, when the heat balance model is trained by using multiple training samples, the heat balance model can determine the graph ontology of the heat balance diagram to be output based on the data of the boundary parameters in the training samples.

[0050] In one implementation, the mapping relationship between the data range of the boundary parameters and the graph ontology can be established in advance according to different types of heat balance diagrams in the basic database, and the heat balance model determines the graph ontology of the heat balance diagram to be output based on the mapping relationship and the input data of the boundary parameters.

[0051] In another implementation, considering that the historical project data is limited and cannot cover all types of heat balance diagrams for the target energy system, in the process of training the heat balance model, the heat balance model needs to learn the mapping relationship between the data of the boundary parameters and the graph ontology.

[0052] In addition, the thermal parameters of each graph node in the graph ontology are obtained based on the data of the boundary parameters and the data of the characteristic parameters. Based on this, in the embodiment of the present application, please refer to Figure 2 , the heat balance model at least includes: a first prediction sub-model and a second prediction sub-model; the first prediction sub-model is used to predict the graph ontology of the heat balance diagram to be output; the second prediction sub-model is used to predict the corresponding thermal parameters for the graph nodes in the predicted graph ontology output by the first prediction sub-model.

[0053] To ensure the training accuracy of the heat balance, in the embodiment of the present application, the first prediction sub-model and the second prediction sub-model are trained by using a joint training method;

[0054] The joint training method includes:

[0055] For each training sample, the data of the boundary parameters in the training sample is input into the first prediction sub-model, and the predicted graph ontology output by the first prediction sub-model is obtained; the predicted graph ontology, the data of the boundary parameters and the data of the characteristic parameters in the training sample are input into the second prediction sub-model, so that the second prediction sub-model outputs the predicted thermal parameters corresponding to each graph node in the predicted graph ontology; the model parameters of the first prediction sub-model and the second prediction sub-model are adjusted by using the first comparison loss and the second comparison loss; the first comparison loss is the comparison loss between the predicted graph ontology and the graph ontology of the heat balance diagram in the training sample, and the second comparison loss is the comparison loss between the predicted thermal parameters and the thermal parameters of each graph node in the heat balance diagram in the training sample.

[0056] In the embodiment of the present application, since the first prediction sub-model and the second prediction sub-model are trained in a joint training manner, and the prediction accuracy of the second prediction sub-model has a strong correlation with whether the prediction graph entity output by the first prediction sub-model is accurate, when adjusting the model parameters, the first prediction sub-model and the second prediction sub-model are taken as a whole, and then the first contrast loss and the second contrast loss are used to jointly adjust the model parameters of the overall heat balance model, so that the trained heat balance model can take into account the prediction of the graph entity and the prediction of the thermal parameters, and the prediction result is more accurate.

[0057] Further, considering the relationship between the graph nodes in the heat balance graph corresponding to the energy system, the second contrast loss can be calculated in the following way:

[0058] Determine the parameter transmission relationship based on the connection relationship of the graph nodes in the predicted graph entity; the parameter transmission relationship includes energy transmission relationship, mass transmission relationship and kinetic energy transmission relationship;

[0059] Output the predicted thermal parameters for each graph node in turn based on the parameter transmission relationship, and calculate the third contrast loss between the predicted thermal parameters of each graph node and the thermal parameters in the training sample, and determine the sum of the third contrast losses corresponding to the plurality of graph nodes as the second contrast loss.

[0060] In the embodiment of the present application, according to the parameter transmission relationship, when predicting the thermal parameters of the graph nodes, the thermal parameters are related to the input energy, input mass, input kinetic energy and output energy, output mass and output kinetic energy, therefore, the thermal parameters of each graph node are predicted in turn according to the parameter transmission relationship, since the thermal parameters of each graph node on the graph entity need to be predicted, the sum of the third contrast losses corresponding to the plurality of graph nodes can be determined as the second contrast loss, which can further improve the prediction accuracy.

[0061] Further, the heat balance model further includes a relationship mapping sub-model for learning the strong correlation mapping relationship between different input parameters; the input parameters include boundary parameters and characteristic parameters;

[0062] Further comprising: when receiving a data adjustment instruction of the input parameters, determining whether the unadjusted parameters need to be adjusted based on the data adjustment result of the input parameters and the strong correlation mapping relationship, and further determining the data of each input parameter input into the second prediction sub-model based on the determination result.

[0063] In the embodiment of the present application, the data of different input parameters have a strong correlation mapping relationship, that is, when the designer inputs the data of the boundary parameters and the data of the characteristic parameters for the first time, when it is necessary to adjust the data of a certain parameter, when the heat balance diagram after adjustment is viewed, due to the strong correlation mapping relationship between the data of different parameters, other parameters corresponding to the strong correlation mapping relationship also need to be adaptively adjusted, at this time, the designer does not need to adjust other parameters, the heat balance model can automatically determine whether the unadjusted parameter needs to be adjusted based on the adjustment result of the data of the input parameter and the strong correlation mapping relationship, if the unadjusted parameter needs to be adjusted, the adjusted data corresponding to the parameter which needs to be adjusted is further determined, and the data of each input parameter input into the second prediction sub-model is further determined based on the determination result, so that the second prediction sub-model can predict the thermal parameters based on the accurate characteristic parameter data, thereby further improving the prediction accuracy.

[0064] In the embodiment of the present application, the heat balance model is obtained by training, even when the working condition of the energy system is changed, the input data of the heat balance model can be adjusted based on the characteristic parameters corresponding to the performance change under different load conditions, so that the corresponding heat balance diagram under the variable working condition can be quickly obtained.

[0065] In one embodiment of the present application, an interactive interface can also be provided, the designer can adjust the input data in real time and view the output heat balance diagram result, thereby realizing efficient, accurate and flexible heat balance diagram calculation, and significantly improving the design efficiency and quality.

[0066] In one embodiment of the present application, the corresponding heat balance model can also be trained for different energy systems, and a guided interface is used to determine the target energy system input by the user, then the heat balance diagram model corresponding to the target energy system is called, the required boundary parameters and characteristic parameters are determined based on the called heat balance diagram model, when the user inputs the corresponding data of the boundary parameters and the characteristic parameters, the corresponding heat balance diagram is output by using the heat balance diagram model.

[0067] The energy system heat balance diagram automatic generation method provided by the embodiment of the present application can reduce the dependence on the experience of designers, reduce human error, significantly improve the calculation efficiency, shorten the design cycle, and reduce the project cost; the heat balance model can provide a more reliable theoretical basis for system design and optimization, and provide system performance and reliability. Moreover, the heat balance model can be widely applied to different energy systems, such as thermal power, compressed air energy storage, gas-steam combined cycle, nuclear power generation, etc., and can realize the design of the system, improve the calculation and analysis accuracy and speed of the energy system of the actual engineering project. In addition, the heat balance diagram generation under the variable working condition can be supported, real-time response to user input and system changes, and suitable for the analysis and optimization of dynamic processes such as system startup, shutdown and load change.

[0068] Please refer to Figure 3 The embodiment of the present application provides an energy system heat balance diagram automatic generation device, which comprises:

[0069] The construction unit 300 is used for constructing a basic database based on historical project data of a target energy system; the basic database comprises a plurality of types of heat balance diagrams corresponding to the target energy system; the heat balance diagram comprises a graph body and thermal parameters marked on a graph node, and the graph body is formed by a plurality of graph nodes and their connection relationships;

[0070] The acquisition unit 302 is used for acquiring a plurality of training samples based on the basic database; the training sample comprises data of boundary parameters and characteristic parameters as input, and further comprises a heat balance diagram as output;

[0071] The training unit 304 is used for training the heat balance model by using a plurality of training samples, so that the heat balance model determines the graph body of the heat balance diagram to be output based on the data of the boundary parameters in the training sample, and predicts the thermal parameters of each graph node in the graph body based on the data of the boundary parameters and the data of the characteristic parameters in the training sample, adjusts the model parameters based on the loss between the prediction result and the thermal parameters of the graph node on the heat balance diagram in the training sample, and obtains the trained heat balance model;

[0072] The heat balance calculation unit 306 is used for outputting a heat balance diagram for input data of a heat balance diagram to be generated by using the heat balance model; the input data is data input according to boundary parameters and characteristic parameters.

[0073] In an embodiment of the present application, the heat balance model at least comprises: a first prediction sub-model and a second prediction sub-model; the first prediction sub-model is used for predicting the graph body of the heat balance diagram to be output; and the second prediction sub-model is used for predicting the corresponding thermal parameters for the graph nodes in the predicted graph body output by the first prediction sub-model.

[0074] In an embodiment of the present application, the first prediction sub-model and the second prediction sub-model are trained in a joint training manner.

[0075] The joint training manner comprises:

[0076] For each training sample, the following steps are performed: inputting the data of the boundary parameters in the training sample into the first prediction sub-model, and obtaining the predicted graph ontology output by the first prediction sub-model; inputting the predicted graph ontology and the data of the boundary parameters and the data of the characteristic parameters in the training sample into the second prediction sub-model, so that the second prediction sub-model outputs the predicted thermal parameters corresponding to each graph node in the predicted graph ontology; and adjusting the model parameters of the first prediction sub-model and the second prediction sub-model together by using a first comparison loss and a second comparison loss; the first comparison loss is a comparison loss between the predicted graph ontology and the graph ontology of the heat balance graph in the training sample, and the second comparison loss is a comparison loss between the predicted thermal parameters and the thermal parameters of each graph node on the heat balance graph in the training sample.

[0077] In an embodiment of the present application, the calculation manner of the second comparison loss comprises: determining a parameter transmission relationship based on the connection relationship of the graph nodes in the predicted graph ontology; the parameter transmission relationship comprises an energy transmission relationship, a mass transmission relationship and a kinetic energy transmission relationship; and sequentially outputting the predicted thermal parameters for each graph node based on the parameter transmission relationship, and calculating a third comparison loss between the predicted thermal parameters of each graph node and the thermal parameters in the training sample, and determining the sum of the third comparison losses of the plurality of graph nodes one by one as the second comparison loss.

[0078] In an embodiment of the present application, the heat balance model further comprises a relationship mapping sub-model configured to learn a strong correlation mapping relationship between data of different input parameters; the input parameters comprise boundary parameters and characteristic parameters.

[0079] The heat balance calculation unit is further configured to: when receiving a data adjustment instruction of the input parameters, determine whether the unadjusted parameters need to be adjusted based on the data adjustment result of the input parameters and the strong correlation mapping relationship, and further determine the data of each input parameter input into the second prediction sub-model based on the determination result.

[0080] In an embodiment of the present application, the target energy system is at least one of the following: a thermal power generation system, a compressed air energy storage system, a gas-steam combined cycle system and a nuclear power generation system.

[0081] It should be noted that: the above embodiment provides the automatic generation device of the energy system heat balance diagram, only the above-mentioned each function module is divided and is illustrated by example, in actual application, the above-mentioned function distribution can be completed by different function modules according to the needs, namely the internal structure of the device is divided into different function modules to complete all or part of the functions described above. In addition, the automatic generation device of the energy system heat balance diagram provided in the above embodiment and the automatic generation method of the energy system heat balance diagram belong to the same concept, and the specific implementation process is described in the method embodiment, which will not be repeated here.

[0082] Embodiments of the present application also provide a computer device, which refers to Figure 4 The computer device includes a processor and a memory, and the memory stores at least one instruction, at least one program, a code set or an instruction set, which is loaded and executed by the processor to implement the automatic generation method of the energy system heat balance diagram provided by the above method embodiments.

[0083] Embodiments of the present application also provide a computer readable storage medium, which stores at least one instruction, at least one program, a code set or an instruction set, which is loaded and executed by the processor to implement the automatic generation method of the energy system heat balance diagram provided by the above method embodiments.

[0084] Embodiments of the present application also provide a computer program product, which includes a computer program, and the processor of the computer device reads the computer program from the computer readable storage medium, and the processor executes the computer program to make the computer device execute the automatic generation method of the energy system heat balance diagram described in any of the above embodiments.

[0085] For the convenience of description, the above system or device is described as various modules or units in function. Of course, in the implementation of the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0086] From the above description of the embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software and necessary general hardware platforms. Based on such understanding, the technical solutions of the present application can be embodied in the form of software products, which can be stored in storage media, such as ROM / RAM, magnetic disks, optical disks, etc., including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments of the present application.

[0087] Finally, it needs to be pointed out that, in this document, relational terms such as first, second, third, and fourth and the like can only be used to distinguish one entity or action from another entity or action, without necessarily requiring or implying any actual relationship or order between or among such entities or actions. Moreover, the terms "comprises", "comprising", or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0088] The above description is only the preferred embodiment of the present application, and it should be pointed out that, for those skilled in the art, some improvements and refinements can be made without departing from the principles of the present application, and these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A method for automatic generation of a heat balance diagram of an energy system, characterized in that, The method comprises: constructing a basic database based on historical project data of a target energy system; the basic database comprises a plurality of types of heat balance diagrams corresponding to the target energy system; the heat balance diagram comprises a graph ontology and thermal parameters marked on graph nodes, the graph ontology is formed by a plurality of graph nodes and their connection relationships; the graph nodes are used to represent components in the target energy system, and the connection relationships are used to represent parameter transmission relationships; the parameter transmission relationships include energy transmission relationships, mass transmission relationships and kinetic energy transmission relationships; based on the basic database, a plurality of training samples are obtained; the training samples include data of boundary parameters and characteristic parameters as input, and also include heat balance diagrams as output; the boundary parameters are external constraints, used to describe input / output limits of the energy system interacting with the outside world; the characteristic parameters are key variables describing inherent properties or dynamic behaviors of the energy system, used to reflect physical, chemical or functional characteristics of the energy system itself; the boundary parameters and the characteristic parameters are determined based on historical project information of corresponding types of energy systems; training a heat balance model using a plurality of training samples, so that the heat balance model determines a graph ontology of a to-be-output heat balance diagram based on data of boundary parameters in the training samples, and predicts thermal parameters of each graph node in the graph ontology based on data of boundary parameters and data of characteristic parameters in the training samples, adjusts model parameters based on a loss between the prediction result and thermal parameters of graph nodes in the heat balance diagram in the training samples, and obtains a trained heat balance model; using the heat balance model to output a heat balance diagram for input data of a to-be-generated heat balance diagram; the input data is input in the form of boundary parameters and characteristic parameters.

2. The method of claim 1, wherein, The heat balance model at least comprises: a first prediction sub-model and a second prediction sub-model; the first prediction sub-model is used to predict a graph ontology of a to-be-output heat balance diagram; the second prediction sub-model is used to predict corresponding thermal parameters for graph nodes in the predicted graph ontology output by the first prediction sub-model.

3. The method of claim 2, wherein, The first prediction sub-model and the second prediction sub-model are trained in a joint training manner; The joint training manner comprises: for each training sample, input data of boundary parameters in the training sample into the first prediction sub-model, and obtain a predicted graph ontology output by the first prediction sub-model; input the predicted graph ontology, data of boundary parameters and data of characteristic parameters in the training sample into the second prediction sub-model, so that the second prediction sub-model outputs predicted thermal parameters corresponding to each graph node in the predicted graph ontology; adjust model parameters of the first prediction sub-model and the second prediction sub-model using a first comparison loss and a second comparison loss; the first comparison loss is a comparison loss between the predicted graph ontology and a graph ontology of a heat balance diagram in the training sample, and the second comparison loss is a comparison loss between the predicted thermal parameters and thermal parameters of each graph node in the heat balance diagram in the training sample.

4. The method of claim 3, wherein, The calculation manner of the second contrast loss comprises: determining a parameter transmission relationship based on the connection relationship of the graph nodes in the predicted graph ontology; outputting the predicted thermal parameters for each graph node in turn based on the parameter transmission relationship, and calculating a third contrast loss between the predicted thermal parameter of each graph node and the thermal parameter in the training sample, and determining the sum of the third contrast losses of the plurality of graph nodes one by one as the second contrast loss.

5. The method of claim 3, wherein, The heat balance model further comprises a relationship mapping sub-model for learning a strong correlation mapping relationship between data of different input parameters; the input parameters comprise boundary parameters and characteristic parameters; Further comprising: when receiving a data adjustment instruction of the input parameters, determining whether the parameters that are not adjusted need to be adjusted based on the data adjustment result of the input parameters and the strong correlation mapping relationship, and further determining the data of each input parameter input into the second prediction sub-model based on the determination result.

6. The method according to any one of claims 1 to 5, characterized in that, The target energy system is at least one of the following: a thermal power generation system, a compressed air energy storage system, a gas-steam combined cycle system, and a nuclear power generation system.

7. An apparatus for automatically generating a heat balance diagram of an energy system, characterized by A device for automatically generating an energy system heat balance diagram as claimed in any one of claims 1-6; the device comprises: a construction unit configured to construct a basic database based on historical project data of a target energy system; the basic database comprises a plurality of types of heat balance diagrams corresponding to the target energy system; the heat balance diagram comprises a graph ontology formed by a plurality of graph nodes and their connection relationship, and thermal parameters labeled on the graph nodes; an acquisition unit configured to acquire a plurality of training samples based on the basic database; the training sample comprises data of boundary parameters and characteristic parameters as input, and comprises a heat balance diagram as output; a training unit configured to train a heat balance model using the plurality of training samples, so that the heat balance model determines a graph ontology of a heat balance diagram to be output based on data of boundary parameters in the training sample, and predicts thermal parameters of each graph node in the graph ontology based on data of boundary parameters and data of characteristic parameters in the training sample, and adjusts model parameters based on a loss between the prediction result and thermal parameters of graph nodes in the heat balance diagram in the training sample, to obtain a trained heat balance model; a heat balance calculation unit configured to output a heat balance diagram for input data of a heat balance diagram to be generated using the heat balance model; the input data is input in the form of boundary parameters and characteristic parameters.

8. A computer device, comprising: The computer device comprises a memory and a processor; the memory is used to store a computer program; and the processor is used to execute the computer program stored in the memory to implement the steps of the method of any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program; and the computer program is executed by the processor to implement the steps of the method of any one of claims 1-6.

10. A computer program product, characterised in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1-6.

Images