WO2021062753A1 - 综合能源系统的仿真方法、装置和计算机可读存储介质 - Google Patents

综合能源系统的仿真方法、装置和计算机可读存储介质 Download PDF

Info

Publication number
WO2021062753A1
WO2021062753A1 PCT/CN2019/109673 CN2019109673W WO2021062753A1 WO 2021062753 A1 WO2021062753 A1 WO 2021062753A1 CN 2019109673 W CN2019109673 W CN 2019109673W WO 2021062753 A1 WO2021062753 A1 WO 2021062753A1
Authority
WO
WIPO (PCT)
Prior art keywords
simulation
model
linear
integrated energy
energy system
Prior art date
Application number
PCT/CN2019/109673
Other languages
English (en)
French (fr)
Inventor
王德慧
江宁
张拓
田中伟
Original Assignee
西门子股份公司
西门子（中国）有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 西门子股份公司, 西门子（中国）有限公司 filed Critical 西门子股份公司
Priority to EP19948148.2A priority Critical patent/EP4024279A4/en
Priority to PCT/CN2019/109673 priority patent/WO2021062753A1/zh
Priority to US17/764,529 priority patent/US20220358268A1/en
Priority to CN201980100606.3A priority patent/CN114503120A/zh
Publication of WO2021062753A1 publication Critical patent/WO2021062753A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2113/00—Details relating to the application field
    • G06F2113/04—Power grid distribution networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/06—Power analysis or power optimisation

Definitions

  • the present invention relates to the technical field of integrated energy, in particular to a simulation method, device and computer readable storage medium of an integrated energy system.
  • Integrated energy service refers to the use of advanced physical information technology and innovative management models in a certain area to integrate multiple energy sources such as coal, oil, natural gas, electric energy, and heat energy in a certain area to achieve coordinated planning among various heterogeneous energy subsystems. Optimize operation, coordinated management, interactive response and complement each other.
  • the comprehensive energy system that can provide comprehensive energy services refers to the organic coordination and optimization of energy generation, transmission and distribution (energy network), conversion, storage, and consumption in the process of planning, construction, and operation.
  • Integrated system of energy production, supply and marketing It is mainly composed of energy supply networks (such as power supply, gas supply, cooling/heating, etc.), energy exchange links (such as CCHP units, generator sets, boilers, air conditioners, heat pumps, etc.), energy storage links (power storage, gas storage, etc.) Heat storage, cold storage, etc.), terminal integrated energy supply units (such as microgrids) and a large number of terminal users.
  • control process for the integrated energy system is cumbersome and highly complex.
  • the embodiment of the present invention proposes a control method, device and computer readable storage medium for an integrated energy system.
  • the process of establishing the simulation model includes: determining the topological structure of the integrated energy system, the topological structure including the equipment of the integrated energy system and the connection attributes between the equipment; determining the general model of the equipment and the corresponding A connector model of connection attributes; connect the general model via the connector model to form a simulation model of the integrated energy system; and train the simulation model.
  • the integrated energy system further includes linear equipment
  • the method also includes the process of pre-generating the general model of each linear device and the process of generating the general model of each nonlinear device in advance, wherein the process of generating the general model of each nonlinear device includes:
  • the ratio of the similarity number supported by the similarity criterion to the similarity number based on the design point data is used to establish the description formula of the non-linear basis process to obtain the general model of the non-linear basis process;
  • the general model includes variable parameters that change non-linearly with the change of actual working conditions;
  • the general model of all the target non-linear general processes of each non-linear device and its associated machine learning algorithm form a general model of this kind of non-linear device.
  • the training simulation model includes:
  • the training of the general model based on the historical data of the device includes: a process of training the general model of the non-linear device based on the historical data of the non-linear device;
  • the process includes:
  • For each target non-linear basic theory process of the non-linear equipment obtain the historical data of actual working condition parameters and variable parameters corresponding to the target non-linear basic theory process of the non-linear equipment, and use the historical data to compare all the parameters.
  • the machine learning algorithm is trained to obtain a variable parameter training model of the target nonlinear basic process
  • variable parameter training model of the target non-linear basis process into the general model of the target non-linear basis process to obtain the trained model of the target non-linear basis process of the non-linear device;
  • the post-training model of all the target non-linear basic processes of the non-linear device constitutes the post-training model of the non-linear device.
  • the non-linear equipment includes: a gas turbine, a heat pump:
  • the target non-linear basic process of the gas turbine includes: the correlation process of flow and pressure in the expansion turbine, and the process of energy conversion between thermal energy and mechanical energy;
  • the target non-linear basic process of the heat pump includes: a heat transfer process, a process of converting heat energy into kinetic energy, a pipeline resistance process, and a process related to flow and pressure.
  • the simulation task includes at least one of the following:
  • Simulation tasks related to equipment performance monitoring ; simulation tasks containing operating assumptions; simulation tasks related to performance monitoring of connector models; simulation tasks related to overall performance monitoring of integrated energy systems.
  • the linear programming algorithm includes at least one of the following:
  • MIP algorithm MIP algorithm
  • MILP algorithm MILP algorithm
  • a simulation device for an integrated energy system including a non-linear device, the device including:
  • the receiving module is used to receive simulation tasks
  • An equation group establishment module which is used to establish a nonlinear equation group based on the simulation task and the simulation model of the integrated energy system;
  • Solving module used to solve nonlinear equations based on linear programming algorithm to obtain simulation results
  • the process of establishing the simulation model includes: determining the topological structure of the integrated energy system, the topological structure including the equipment of the integrated energy system and the connection attributes between the equipment; determining the general model of the equipment and the corresponding A connector model of connection attributes; connect the general model via the connector model to form a simulation model of the integrated energy system; and train the simulation model.
  • the integrated energy system further includes linear equipment
  • the process of establishing the simulation model also includes the process of generating a general model of each linear device and nonlinear device in advance, and the process of generating a general model of each nonlinear device includes: for each target of each nonlinear device Determine the complete design point data for the non-linear basis process; adopt the ratio of the similarity number supported by the similarity criterion to the similarity number based on the design point data to establish the description formula of the non-linear basis process to obtain the non-linearity
  • the general model of the basic process; the general model of the non-linear basic process includes variable parameters that change nonlinearly with the change of actual working condition parameters; constructing the relationship between the actual working condition parameters and the variable parameters Machine learning algorithm, and establish the association relationship between the machine learning algorithm and the general model of the non-linear basic process; the general model of the non-linear general process of all the targets of each non-linear device and its associated machine
  • the learning algorithm constitutes a general model of this kind of nonlinear equipment.
  • the simulation task includes at least one of the following:
  • Simulation tasks related to equipment performance monitoring ; simulation tasks containing operating assumptions; simulation tasks related to performance monitoring of connector models; simulation tasks related to overall performance monitoring of integrated energy systems.
  • the linear programming algorithm includes at least one of the following:
  • MIP algorithm MIP algorithm
  • MILP algorithm MILP algorithm
  • a simulation device for an integrated energy system including a processor and a memory
  • An application program executable by the processor is stored in the memory, and is used to make the processor execute the simulation method of the integrated energy system as described in any one of the above.
  • a computer-readable storage medium in which computer-readable instructions are stored, and the computer-readable instructions are used to execute the simulation method of an integrated energy system as described in any one of the above.
  • a simulation task is received; a nonlinear equation system is established based on the simulation task and the simulation model of the integrated energy system; the nonlinear equation system is solved based on the linear programming algorithm to obtain the simulation result; the simulation result is output;
  • the establishment process of the simulation model includes: determining the topological structure of the integrated energy system, the topological structure including the equipment of the integrated energy system and the connection attributes between the equipment; determining the general model of the equipment and the corresponding connection attributes A connector model; connect the general model via the connector model to form a simulation model of the integrated energy system; train the simulation model.
  • the equipment-based general model and the connector model constitute a simulation model of the integrated energy system; the simulation model is trained, and the control commands of the integrated energy system are generated based on the trained simulation model. Therefore, through the reusable universal model, the sharing of professional knowledge among various industries and users is realized, which facilitates the construction of a simulation system of an integrated energy system, and improves the control efficiency of the integrated energy system.
  • the implementation mode modifies the model parameters of the equipment in a training manner according to actual operating conditions, so as to achieve higher accuracy.
  • the ratio of the similarity number supported by the similarity criterion to the similarity number based on the design point data is used to establish the description formula of the non-linear basis process to obtain the non-linear basis
  • the general model of the process makes the model applicable to a class of equipment.
  • the general model includes variable parameters that change non-linearly with the actual working condition parameters, and by constructing a machine learning algorithm between the actual working condition parameters and the variable parameters, the variable parameters can be learned through machine learning. Obtained, which in turn enables the passing model to have self-learning capabilities.
  • the target nonlinear basis process obtains the actual working condition parameters and the historical data of variable parameters corresponding to the target non-linear basis process of the specific equipment, and use The historical data trains the machine learning algorithm to obtain the variable parameter training model of the target nonlinear basic process, and substitute the variable parameter training model of the target nonlinear basic process into the general model of the target nonlinear basic process.
  • Obtain the trained model of the target non-linear basic process of the specific equipment that is, the instantiated model that meets the characteristics of the specific equipment.
  • the non-linear model can be made available when the training variable parameters are not available on site, for example, when there is insufficient historical data.
  • Fig. 1 is a flowchart of a control method of an integrated energy system according to an embodiment of the present invention.
  • Fig. 2 is a schematic diagram of a simulation model of an integrated energy system based on a connector model and a general model of equipment in an embodiment of the present invention.
  • Fig. 3 is a schematic diagram of setting parameters of a general model of a device in an embodiment of the present invention.
  • Fig. 4 is a flowchart of a method for generating a general model of a non-linear device according to an embodiment of the present invention.
  • Fig. 5 is a block diagram of a control device of an integrated energy system according to an embodiment of the present invention.
  • FIG. 6 is a structural diagram of a control device of an integrated energy system with a processor-memory architecture according to an embodiment of the present invention.
  • Fig. 7 is an exemplary topological structure diagram of an integrated energy system according to an embodiment of the present invention.
  • FIG. 8 is a flowchart of a simulation method of an integrated energy system according to an embodiment of the present invention.
  • Fig. 9 is a structural diagram of a simulation device of an integrated energy system according to an embodiment of the present invention.
  • FIG. 10 is a structural diagram of a simulation device of an integrated energy system with a processor-memory architecture according to an embodiment of the present invention.
  • Fig. 11 is a flowchart of a method for optimizing an integrated energy system according to an embodiment of the present invention.
  • Fig. 12 is a structural diagram of an optimization device of an integrated energy system according to an embodiment of the present invention.
  • FIG. 13 is a structural diagram of an optimization device for an integrated energy system with a processor-memory architecture according to an embodiment of the present invention.
  • FIG. 14 is a system structure diagram of a simulation example of an integrated energy system according to an embodiment of the present invention.
  • FIG. 15 is a system structure diagram of an optimized example of the integrated energy system according to the embodiment of the present invention.
  • the reusable equipment universal model is used to form a simulation model of the integrated energy system, and the equipment’s universal model is trained based on the historical data of the equipment, thereby realizing the training of the simulation model and generating it from the trained simulation model Integrated energy system control commands, thereby reducing control complexity.
  • Fig. 1 is a flowchart of a control method of an integrated energy system according to an embodiment of the present invention.
  • the method includes:
  • Step 101 Determine the topological structure of the integrated energy system.
  • the topological structure includes the equipment of the integrated energy system and the connection attributes between the equipment.
  • the integrated energy system is determined The topology.
  • the topological structure includes the equipment of the integrated energy system and the connection attributes between each equipment.
  • the equipment included in the integrated energy system may include: absorption refrigeration (AC) equipment, multi-effect evaporation (MED) equipment, compression refrigeration (CC) equipment, multi-stage flash distillation (MSFD) equipment, combined heat and Cold supply (CHC) equipment, solar photovoltaic (PV) equipment, cold water energy storage (CWS) equipment, cold and hot convertible heat pump (rHP) equipment, electric boiler (EB) equipment, reverse osmosis (RO) equipment, electrochemical energy storage (ECES) equipment, steam turbine (ST) equipment, gas boiler (GB) equipment, water storage (WS) equipment, gas turbine (GT) equipment, wind power (WT) equipment, gas turbine inlet cooler (HEX) equipment, internal combustion engine (ICE) Equipment, heat pump (HP) equipment, ice storage-integrated desalination (isiD) equipment, high temperature heat storage/low temperature heat storage (HTS/LTS) equipment, low temperature heat storage (ITES) equipment, ice thermal energy storage (ITES) equipment, and many more.
  • AC absorption refrigeration
  • MED
  • connection attributes between devices may include:
  • connection characteristics between devices for example, after determining the connection relationship between the devices, the connection characteristics specifically include pipe connections, electrical connections, economic connections, and so on.
  • Fig. 7 is an exemplary topological structure diagram of an integrated energy system according to an embodiment of the present invention.
  • the topology of the integrated energy system includes an energy source 701, an energy conversion part 702, an energy consumption part 703, and an energy form 704 at the end user. among them:
  • Energy 701 includes: natural gas 70, PV/WIND 71 and solar thermal energy 72.
  • the energy conversion part 702 includes: high temperature heat storage 73, compression heat pump 74, absorption heat pump 75, gas turbine 76, first gas boiler 77, second gas boiler 78, first steam generator 79, second steam generator 80, back pressure (back press) Turbine 81, medium temperature heat storage 82, compressed air conditioning 83, absorption conditioning 84, LP boiler 85, cold storage 86 and low temperature heat storage 87.
  • the energy consumption part 703 includes: process power 88, in-process flow 89, water treatment 90, cold water island 91, cooling process 92, compressed air 93 and heat treatment/casting/welding 94.
  • the energy forms at the end users include: grid 95, steam 96, refrigeration 97 and heating 98.
  • connection properties between the devices such as the pipe connection between the high temperature heat storage 73 and the second steam generator 80, or the electrical connection between the process power 88 and the grid 95, and so on.
  • Step 102 Determine the general model of the device and the connector model corresponding to the connection attribute.
  • the devices in the topology usually include linear devices and non-linear devices.
  • the method also includes a process of pre-generating a general model of each linear device and a general model of each non-linear device.
  • the modeling process of the linear device can adopt general linear or multiple regression, etc., which will not be repeated in the present invention.
  • the process of generating a general model of each non-linear device includes:
  • each target non-linear basis process of each non-linear device determine its complete design point data; adopt the ratio of the similarity number supported by the similarity criterion to the similarity number based on the design point data to establish the non-linear basis theory
  • the description formula of the process obtains the general model of the non-linear basis process; the general model of the non-linear basis process includes variable parameters that change nonlinearly with the change of the actual working condition parameters; constructing the actual working condition
  • the machine learning algorithm between the parameters and the variable parameters, and establish the association relationship between the machine learning algorithm and the general model of the nonlinear basic process; make all the targets of each nonlinear device non-linear universal
  • the general model of the process and its associated machine learning algorithms constitute a general model of this kind of nonlinear equipment.
  • the connector model can be implemented as a wire model, a pipe connection line model, an economics connection line model, and so on.
  • Step 103 Connect the universal model via the connector model to form a simulation model of the integrated energy system.
  • the universal model can be connected via the connector model through drag and drop on the graphical interface to form a simulation model of the integrated energy system.
  • the additional input parameters can also be marked in the form of a visual arrow.
  • Step 104 Training the simulation model.
  • training the simulation model includes: obtaining historical data of each device in the integrated energy system during the operation of the simulation model; training a general model based on the historical data of these devices.
  • the equipment includes linear equipment and non-linear equipment
  • the historical data of the linear equipment trains the general model of the linear equipment
  • the general model of the non-linear equipment is trained based on the historical data of the non-linear equipment.
  • the user can freely define the object target to be trained and the corresponding decision parameters, and freely select the training algorithm, the corresponding historical data, and the appropriate technical parameters.
  • the parameters of the model can be entered manually, or can be optimized automatically based on stored long-term historical data and real-time data to obtain a model with higher accuracy.
  • the trained data model and mechanism model are the data basis for subsequent prediction and optimization.
  • the process of training the general model of the non-linear device includes: for each target non-linear basis process of the non-linear device, obtaining the actual working condition parameters corresponding to the target non-linear basis process of the non-linear device and Variable parameter historical data, using the historical data to train the machine learning algorithm to obtain a variable parameter training model of the target non-linear basis process; convert the variable parameters of the target non-linear basis process
  • the training model is substituted into the general model of the target non-linear basis process to obtain the trained model of the target non-linear basis process of the non-linear device; and all the target non-linear basis processes of the non-linear device are obtained.
  • the post-training model of the non-linear device constitutes the post-training model of the nonlinear device.
  • the non-linear equipment includes: a gas turbine and a heat pump: the target non-linear basic process of the gas turbine includes: the process of correlation between flow and pressure in the expansion turbine, and the process of energy conversion between thermal energy and mechanical energy.
  • the target non-linear basic process of the heat pump includes: a heat transfer process, a process of converting heat energy into kinetic energy, a pipeline resistance process, and a process related to flow and pressure.
  • Step 105 Generate a control command of the integrated energy system based on the training simulation model.
  • control command of the integrated energy system can be generated based on the training simulation model.
  • the control command can be a control command for a single device in the integrated energy system, or a comprehensive control command for the integrated energy system.
  • control commands can be implemented as simulation commands or optimization commands. For example, receiving an optimization task containing an optimization target; inputting the optimization target into the simulation model; enabling the simulation model to output control commands that meet the constraints in the simulation model and reach the optimization target.
  • the method further includes: receiving a simulation task; running the simulation model based on the simulation task and outputting a simulation result; wherein the simulation task includes at least one of the following: a simulation task related to equipment performance monitoring; including operations The simulation task of hypothetical conditions; the simulation task of the performance monitoring of the connector model; the simulation task of the overall performance monitoring of the integrated energy system, etc.
  • the simulation task specifically includes the performance monitoring of the integrated energy system, energy-saving management, and system parameter optimization operation.
  • Optimization tasks include: optimal scheduling to meet a certain period of load within a certain period of time.
  • the process shown in Figure 1 can be integrated into a software platform, and the process shown in Figure 1 can be implemented based on the software platform.
  • the software platform can follow the general IOT platform architecture, provide microservices, and build the PAAS layer and SAAS layer.
  • the front end is connected to the SCADA of the local data storage layer, or connected to other systems through OPC UA to obtain data.
  • the platform system has the function of parsing the MQTT Internet of Things protocol. Each site can convert different communication protocols into MQTT protocols and then upload them.
  • Platform for data communication follow the IEC61970 and IEC61850 standard data information model standards, so as not only to realize the transmission of the data itself, but also to freely suggest or receive the semantic relations of various types of data, such as the hierarchical relationship, system connection, business logic, and so on.
  • the various algorithm functions of the integrated energy service are closely related, and can run independently, interact with each other, and work together to complete a variety of energy management tasks, such as: integrated energy-related component modules, big data Analysis modules, system modules, optimization modules, simulation modules, etc., are all carried on the container technology, support distributed layout, support cross-platform, and support cloud platform.
  • IEMS uses MYSQL database and RABBITMQ to form message and data distribution bus technology, which is used for message transmission, data transmission and task distribution between algorithms and between algorithms and the front end.
  • CELERY is used for message communication and business coordination between DOCKER.
  • the ADAPTOR layer is used to communicate with the SIEMENS SCADA system or SIEMENS DEOP, or communicate with other systems and devices that follow the OPC UA protocol, or directly receive the MQTT protocol.
  • PaaS layer services include services related to the functions of the integrated energy management and control system: data processing services, component model library services, model training services, energy system construction services, prediction services, optimization services, report services, and dashboard user interface services, Data analysis tool services; general PaaS layer services related to information and communications include: message reception and distribution services, load balancing services, data management services, etc.
  • SaaS services realize: energy network data and performance monitoring and diagnosis, short-term forecasting of load and power generation, system parameter operation optimization, optimal scheduling management of energy production and supply, factory peak energy management, energy saving management, multi-network coordination management, predictability Maintenance management.
  • the software platform can include a simulation environment and an optimization environment.
  • the simulation environment refers to the state of the system at a certain moment. It can be connected to the state analyzed by the measurement points below, or it can be analyzed by adding some operating assumptions.
  • operation optimization refers to the optimal overall operation in a certain period of time, which is a period of time. They have simulation solvers and optimization solvers for each word, and their own intermediate interpreters. They share a scene and problem to establish a user graphical interface as a semantic input system for equipment, measuring points, positions, connection relations, etc., and use their own interpreters. Form the mathematical problems received by the respective algorithms and solve them.
  • the embodiment of the present invention inputs the boundary conditions and design goals of various energy sources in the park, and forms a reusable standard model through data processing, model training, etc., and at the same time generates a standard model that matches the historical load.
  • the unit price of electricity, heat, and gas can be reduced, the equipment cost of each energy module can be reduced, and the absorption capacity of renewable energy can be improved, which can effectively improve the economy of the system.
  • the historical data and the real-time latest data are integrated, the corresponding model is revised, and various advanced algorithms are used to form a real-time operation scheduling strategy to ensure that the system can always reach the setting during operation.
  • the optimal goal Provide a reliable information technology foundation for subsequent real-time power market transactions, demand response, and other energy market real-time transactions.
  • the software platform can also realize the visualization of industrial energy data and the transparent display of data.
  • big data analysis strategy based on mechanism model and artificial intelligence technology, real-time prediction of user load, real-time dynamic optimization of short-term operation of power generation forecasting, to cope with the current market peak and valley filling, step electricity price, frequency regulation and voltage regulation, energy saving and consumption reduction in factories And future free market trading (futures and spot trading, the previous day and the day trading market) demand.
  • the software platform can be optimized for operation and management. Monitor the mid-term performance and status of each component in the system, and conduct real-time energy efficiency analysis. Use real-time collected energy data to conduct simulation analysis, formulate technical transformation plan planning, and technical and economic analysis.
  • the software platform can make full use of renewable energy, fossil fuels, residual temperature and pressure, new energy and other resource forms to coordinate with each other, through the flexible operation of the network and storage, establish innovative business mechanisms, and adopt intelligent Means to realize the comprehensive supply of high-quality, high-efficiency and economical regional power, heating, cooling, gas and other loads, to meet the requirements of random changes in terminal loads.
  • the optimization algorithm automatically optimizes the scheduling of the controllable resources through the lower-level control system, and accurately reflects the effects of the uncontrollable resources.
  • the system can automatically recognize the changes of the system, so as to realize automatic and flexible changes.
  • the platform can choose optimization goals, including energy costs, carbon dioxide emissions, and efficiency.
  • Fig. 2 is a schematic diagram of a simulation model of an integrated energy system based on a connector model and a general model of equipment in an embodiment of the present invention.
  • the gas turbine general model 202, the lithium bromide heating machine general model 213, the lithium bromide refrigerator general model 214, the power supply load general model 215, the heating load general model 216, and the cooling load general model 217 have all been established in advance. Then, by dragging, the gas turbine general model 202, the lithium bromide heating machine general model 213, and the lithium bromide refrigerator general model 214 are connected to each other to form a simulation model as shown in FIG. 2.
  • the connector model can contain arrows and connections.
  • the arrow-based can provide input for the general model; the wire-based can provide the electrical connection, pipe connection or economic connection for the general model.
  • the arrow 201, the arrow 210, the arrow 205, the arrow 206, the arrow 207, the arrow 208, and the arrow 209 can respectively provide input for the general model pointed by the arrow.
  • Fig. 3 is a schematic diagram of setting parameters of a general model of a device in an embodiment of the present invention.
  • the parameter setting area 220 is further displayed.
  • the parameter input boxes related to the gas turbine general model 202 are displayed, for example, the temperature input box 221, the pressure ratio input box 222, the pressure recovery input box 223, the inlet temperature input box 224, and the air flow input box 225 are displayed. . Then, the user can respectively input parameters in these parameter input boxes, and these parameters are assigned to the gas turbine general model 202.
  • the embodiment of the present invention also proposes a method for generating a general model of a non-linear device.
  • Fig. 4 is a flowchart of a method for generating a general model of a non-linear device according to an embodiment of the present invention.
  • the method includes:
  • Step 401 For each target non-linear basis process of the non-linear device, obtain historical data of actual working condition parameters and variable parameters corresponding to the target non-linear basis process of the non-linear device.
  • the basic process can sometimes be called a physical process, such as the heat transfer process, the electrical energy conversion process, and the aforementioned flow and pressure related processes.
  • the basic process of interest can be determined, that is, the basic process that needs to be modeled.
  • These basic processes that need to be modeled are called the target basic process, and the nonlinear target basic process is It can be called the target nonlinear basis process.
  • the target non-linear basic process can include: the process of flow and pressure in the expansion turbine, the process of energy conversion between thermal energy and mechanical energy, etc.; for the heat pump, the target non-linear basic process can include: Thermal process, the process of converting heat energy into kinetic energy, the pipeline resistance process, the process of flow and pressure, etc.
  • the complete design point data can be restored based on the published design parameters and the general design point information provided by the manufacturer to the user.
  • its design point data can include pressure ratio, air flow, etc. Based on these design point data, it can derive its efficiency, inlet resistance, and air extraction volume and other related designs that are not available to users. parameter.
  • Step 402 Use the ratio of the similarity number supported by the similarity criterion to the similarity number based on the design point data to establish a description formula of the non-linear basis process to obtain a general model of the non-linear basis process; the general model It includes variable parameters that change non-linearly with changes in actual operating conditions.
  • the actual working condition parameters refer to the parameters of a specific equipment that change with the actual working condition parameters.
  • the variable parameter may have a preset default value.
  • G1 is the flow rate
  • T1 is the temperature
  • P1 is the pressure
  • G0 is the corresponding design point flow
  • T0 is the corresponding design point temperature
  • P0 is the corresponding design point pressure
  • f() is a function
  • coefficients a and b are variable parameters that change non-linearly with actual operating conditions.
  • a default value can also be set for the variable parameters a and b.
  • IGV is the angle of the imported transmissible vane.
  • Step 403 Construct a machine learning algorithm between the actual working condition parameters and the variable parameters.
  • a machine learning algorithm between the actual working condition parameter and the variable parameter can be constructed based on a machine learning big data analysis method such as a neural intelligent network or a support vector machine
  • Step 404 The general model of the non-linear basic process of all targets of each type of equipment and its associated machine learning algorithm constitute a general model of the type of equipment.
  • the training process of the general model includes:
  • the first step for each target non-linear basis process of a specific equipment of this kind of equipment, obtain the historical data of actual working condition parameters and variable parameters corresponding to the target non-linear basis process of the specific equipment, The historical data is used to train the corresponding machine learning algorithm to obtain the variable parameter training model of the target nonlinear basic process.
  • a set of historical data of actual working condition parameters are used as input sample values, and the historical data of variable parameters corresponding to the set of historical data of actual working condition parameters are used as output sample values.
  • the machine learning algorithm is trained on the input sample value and the corresponding output sample value of, and the self-learning model with variable parameters, also called training model, can be obtained.
  • the historical data of the relevant actual working condition parameters of the gas turbine on site and the historical data of the corresponding variable parameters can be obtained, and the input and output sample sets can be obtained.
  • a training model with variable parameters a and b can be obtained.
  • Step 2 Substitute the variable parameter training model of the target non-linear basis process into the general model of the target non-linear basis process to obtain the training model of the target non-linear basis process of the specific device model.
  • the trained model is a self-learning model with learning ability.
  • variable parameter training model of the target nonlinear basis process can be substituted into the general model of the target nonlinear basis process.
  • the third step the trained models of all target non-linear basic processes of the specific equipment constitute the post-training model of the specific equipment.
  • the input parameters of the trained model may include all input parameters required by the trained model of the target nonlinear basis process.
  • the modeling method of the nonlinear model in the embodiment of the present invention has been described in detail above.
  • the nonlinear model is usually more complicated in the modeling process, but the corresponding accuracy is also relatively high.
  • the nonlinear model is directly used to run the corresponding control (such as simulation or optimization) of the integrated energy system, it may be due to the nonlinear model.
  • the operating speed is relatively slow, which affects the real-time simulation of the entire integrated energy system.
  • the first step for the nonlinear model of each device, determine the value range of each input parameter of the model.
  • the effective power range is 50% to 110% of the rated operating conditions.
  • Step 2 Divide the value range of each input parameter into multiple subintervals based on multiple interpolation points.
  • the value range of each input parameter can be divided into a plurality of sub-intervals based on a plurality of interpolation points.
  • the interpolation points can be determined by setting more interpolation points for regions with severe nonlinear changes, and setting fewer interpolation points for regions with slow nonlinear changes.
  • 40 points are inserted for the power range, 20 points for the ambient temperature, 5 points for the ambient pressure, etc.
  • Step 3 Determine a plurality of input sample values equally within each sub-interval.
  • a plurality of input sample values can be equalized and determined in each sub-interval. For example, power and ambient temperature can be divided into equal parts in the actual domain.
  • the fourth step traverse the input sample value combination of each input parameter of the model, and use the nonlinear model to obtain the output sample value combination corresponding to each input sample value combination.
  • Step 5 Generate a tensor table using all input sample value combinations and their corresponding output sample value combinations.
  • the efficiency value can be obtained by interpolation.
  • the tensor table is searched according to the current value of each input parameter, and the corresponding data found from the tensor table is used for interpolation processing to obtain the corresponding output value.
  • the current value of each input parameter may be a real value or a hypothetical value.
  • the interpolation algorithm can be selected according to the actual situation, for example, a linear interpolation method or a nonlinear interpolation method can be selected.
  • linear interpolation may be used for points that are close to each other, and nonlinear interpolation may be used for points that are far apart.
  • the value of other output variables corresponding to the required temperature, pressure, and performance can be obtained through methods such as three-dimensional temperature, pressure, and power spline interpolation.
  • This general method uses a general program, regardless of any specific model, such as heat pumps, internal combustion engines, heat exchangers, etc., can be implemented by this method.
  • dividing the value range of each input parameter into a plurality of sub-intervals based on a plurality of interpolation points is: based on an equalization criterion, dividing the value range of each input parameter into a plurality of sub-intervals based on the plurality of interpolation points.
  • the plurality of input sample values are determined equally in each sub-interval: based on an equalization criterion, the plural input sample values are determined in each sub-interval equally.
  • interpolation processing is performed on the tensor table according to the current value of each input parameter to obtain the corresponding output value.
  • the embodiment of the present invention also proposes a control device for an integrated energy system.
  • Fig. 5 is a block diagram of a control device of an integrated energy system according to an embodiment of the present invention.
  • control device 500 includes:
  • the topological structure determining module 501 is configured to determine the topological structure of the integrated energy system, the topological structure including the devices of the integrated energy system and the connection attributes between the devices;
  • the model determination module 502 is configured to determine a general model of the device and a connector model corresponding to the connection attribute;
  • the simulation model composition module 503 is configured to connect the general model via the connector model to form a simulation model of the integrated energy system
  • the training module 504 is used to train the simulation model
  • the control command generation module 505 is configured to generate a control command of the integrated energy system based on the training simulation model.
  • the device includes a linear device and a non-linear device
  • the model determination module 502 is also used for the process of generating a general model of each type of linear equipment and non-linear equipment in advance, wherein the process of generating a general model of each type of non-linear equipment includes: for each target non-linear device of each type of non-linear equipment Based on the basic theory process, determine its complete design point data; adopt the ratio of the similarity number supported by the similarity criterion to the similarity number based on the design point data, establish the description formula of the nonlinear basic theory process, and obtain the nonlinear basic theory
  • the general model of the process; the general model of the non-linear basic process includes variable parameters that change nonlinearly with the change of the actual working condition parameters; constructing machine learning between the actual working condition parameters and the variable parameters Algorithm, and establish the association relationship between the machine learning algorithm and the general model of the non-linear basic process; the general model of the non-linear general process and the associated machine learning algorithm of all the targets of each non-linear device , Constitute a
  • the training module 504 is configured to obtain historical data of the device during the operation of the simulation model; and train the general model based on the historical data of the device.
  • the training of the general model based on the historical data of the device includes: a process of training the general model of the non-linear device based on the historical data of the non-linear device;
  • the training module 504 is configured to obtain historical data of actual working condition parameters and variable parameters corresponding to the target non-linear basis process of the non-linear equipment for each target non-linear basis process of the non-linear equipment , Using the historical data to train the machine learning algorithm to obtain a variable parameter training model of the target nonlinear basic process; substituting the variable parameter training model of the target nonlinear basic process into the target In the general model of the nonlinear basic process, the trained model of the target nonlinear basic process of the nonlinear device is obtained; the trained model of all the target nonlinear basic processes of the nonlinear device is formed to form The post-training model of the non-linear device.
  • the non-linear equipment includes: a gas turbine, a heat pump:
  • the target non-linear basic process of the gas turbine includes: the correlation process of flow and pressure in the expansion turbine, and the process of energy conversion between thermal energy and mechanical energy; or
  • the target non-linear basic process of the heat pump includes: a heat transfer process, a process of converting heat energy into kinetic energy, a pipeline resistance process, and a process related to flow and pressure.
  • control command generation module 505 is further configured to receive a simulation task after the training simulation model; run the simulation model based on the simulation task and output the simulation result; wherein the simulation task Contains at least one of the following: simulation tasks related to equipment performance monitoring; simulation tasks containing operational assumptions; simulation tasks related to performance monitoring of connector models; simulation tasks related to overall performance monitoring of integrated energy systems.
  • control command generation module 505 is configured to receive an optimization task including optimization goals and constraints; input the constraints and the optimization goals into the simulation model; enable the simulation model output to meet all requirements.
  • the control commands of the optimization target are described.
  • the embodiment of the present invention also proposes a control device of an integrated energy system with a processor and a memory structure.
  • Fig. 6 is a structural diagram of a control device of an integrated energy system with a processor-memory structure according to an embodiment of the present invention.
  • control device 600 of the integrated energy system includes a processor 601 and a memory 602;
  • An application program that can be executed by the processor 601 is stored in the memory 602 to enable the processor 601 to execute the control method of the integrated energy system as described in any of the above items.
  • the memory 602 can be specifically implemented as a variety of storage media such as electrically erasable programmable read-only memory (EEPROM), flash memory (Flash memory), and programmable program read-only memory (PROM).
  • the processor 601 may be implemented to include one or more central processing units or one or more field programmable gate arrays, where the field programmable gate array integrates one or more central processing unit cores.
  • the central processing unit or central processing unit core may be implemented as a CPU or MCU.
  • the simulation model platform of the integrated energy system mainly includes a simulation environment and an optimized environment.
  • the simulation environment refers to the state of the integrated energy system at a certain moment, which can be connected to the state analyzed by the measurement points below, or it can be Add some operational hypotheses to analyze the state.
  • the operation optimization environment refers to the optimal overall operation in a certain period of time, which is a period of time.
  • the embodiment of the present invention also realizes the simulation method and optimization calculation method based on the simulation model of the integrated energy system, the simulation method realizes the performance monitoring of the integrated energy system, energy-saving management, and system parameter optimization operation; the optimization method is realized within a certain period of time To meet the optimal scheduling of a certain period of load.
  • FIG. 8 is a flowchart of a simulation method of an integrated energy system according to an embodiment of the present invention.
  • the method 800 includes:
  • Step 801 Receive a simulation task.
  • simulation tasks for equipment performance monitoring For example, receiving from users: simulation tasks for equipment performance monitoring; simulation tasks including operating assumptions; simulation tasks for performance monitoring of connector models; simulation tasks for overall performance monitoring of integrated energy systems. and so on.
  • Step 802 Based on the simulation task and the simulation model of the integrated energy system, a nonlinear equation system is established.
  • a nonlinear equation set is established based on the power balance formula and performance constraint relationship in the simulation model.
  • Step 803 Solve the system of nonlinear equations based on the linear programming algorithm to obtain simulation results.
  • the linear programming algorithm includes at least one of the following: a mixed integer programming (Mixed Integer Programming, MIP) algorithm; a mixed integer linear programming (Mixed Integer Linear Programming, MILP) algorithm, and so on.
  • MIP Mixed integer programming
  • MILP Mixed integer linear programming
  • Step 804 Output the simulation result; wherein the establishment process of the simulation model includes: determining the topological structure of the integrated energy system, the topological structure including the equipment of the integrated energy system and the connection attributes between the equipment; A general model of the equipment and a connector model corresponding to the connection attribute; connecting the general model via the connector model to form a simulation model of the integrated energy system; training the simulation model.
  • the nonlinear model in the simulation model has a tensor table after being trained and linearized.
  • the tensor table of the nonlinear model can be searched according to the current value of the input parameters of each nonlinear model, and the corresponding value found from the tensor table can be used Perform interpolation processing to obtain the corresponding output value, which is the output value of the nonlinear model, thereby realizing the linear programming algorithm to solve the nonlinear equation system.
  • Fig. 9 is a structural diagram of a simulation device of an integrated energy system according to an embodiment of the present invention.
  • the simulation device 900 of the integrated energy system includes:
  • the receiving module 901 is used to receive simulation tasks
  • the equation system establishment module 902 is configured to establish a nonlinear equation system based on the simulation task and the simulation model of the integrated energy system;
  • the solving module 903 is used to solve the system of nonlinear equations based on the linear programming algorithm to obtain simulation results;
  • the output module 904 is configured to output the simulation results; wherein the establishment process of the simulation model includes: determining the topological structure of the integrated energy system, the topological structure including the equipment of the integrated energy system and the connection attributes between the equipment; Determine a general model of the device and a connector model corresponding to the connection attribute; connect the general model via the connector model to form a simulation model of the integrated energy system; and train the simulation model.
  • FIG. 10 is a structural diagram of a simulation device of an integrated energy system with a processor-memory architecture according to an embodiment of the present invention.
  • the simulation device 1000 of the integrated energy system includes a processor 1001 and a memory 1002; the memory 1002 stores an application program that can be executed by the processor 1001, and is used to make the processor 1001 execute any of the above Simulation method of integrated energy system.
  • the following describes an exemplary simulation process of the integrated energy system based on the embodiment of the present invention.
  • FIG. 14 is a system structure diagram of a simulation example of an integrated energy system according to an embodiment of the present invention.
  • the exemplary structure of the simulation model of the integrated energy system shown in Fig. 14 includes a gas turbine model after training and linearization processing, an energy storage battery model, a photovoltaic model, and three load models.
  • the gas turbine supplies power to load 1 and load 2 at the same time, and the photovoltaic system and the grid jointly supply power to load 3.
  • the variables marked in Figure 14 represent the power provided or consumed by different devices.
  • the output of the gas turbine is X out,1 ;
  • the input of load 1 is X in,1 ;
  • the input of load 2 is X in,2 ;
  • the output of the power grid is X out,2 ;
  • the output of the photovoltaic system is X out,3 ;
  • the input of the energy storage battery model is X in,4 ;
  • the output of the energy storage battery model is X out,4 ;
  • the input of load 3 is X in,3 and X out,4 .
  • the embodiments of the present invention can be used to perform simulation calculations on each device and the overall network.
  • the following aspects need to be considered:
  • the optimization goal can be set to the number that should be zero on the right side of the equation as a variable, and the sum of squares of all variables is the smallest, and Turn the balance problem into an optimization problem and use various optimization solvers to solve it.
  • the MILP algorithm By calling the MILP algorithm, the status of each device in the system and the energy flow parameters in the network can be obtained in the current state. And has the following two functions: (1), compare and correct with the sensor in the system. (2) For devices that are not easy to deploy sensors, the calculation results can be used to understand their operating status.
  • the simulation solution function can also know in advance the impact of changes (including parameter changes) in a certain device in the system.
  • Fig. 11 is a flowchart of a method for optimizing an integrated energy system according to an embodiment of the present invention.
  • the method 1100 includes:
  • Step 1101 Receive an optimization task containing an optimization target.
  • Step 1102 Based on the optimization goal and the simulation model of the integrated energy system, a nonlinear equation system is established.
  • Step 1103 Solve the system of nonlinear equations based on the linear programming algorithm to obtain optimization results.
  • Step 1104 Output the optimization result; wherein the establishment process of the simulation model includes: determining the topological structure of the integrated energy system, the topological structure including the equipment of the integrated energy system and the connection attributes between the equipment; determining the A general model of the equipment and a connector model corresponding to the connection attribute; the general model is connected via the connector model to form a simulation model of the integrated energy system; and the simulation model is trained.
  • the nonlinear model in the simulation model has a tensor table after being trained and linearized.
  • the tensor table of the nonlinear model can be searched according to the current value of the input parameters of each nonlinear model, and the corresponding value found from the tensor table can be used Perform interpolation processing to obtain the corresponding output value, which is the output value of the nonlinear model, thereby realizing the linear programming algorithm to solve the nonlinear equation system.
  • Fig. 12 is a structural diagram of an optimization device of an integrated energy system according to an embodiment of the present invention.
  • the optimization device 1200 of the integrated energy system includes:
  • the receiving module 1201 is used to receive optimization tasks including optimization targets;
  • the equation system establishment module 1202 is configured to establish a nonlinear equation system based on the optimization target and the simulation model of the integrated energy system;
  • the solving module 1203 is used to solve the nonlinear equation system based on the linear programming algorithm to obtain the optimization result;
  • the output module 1204 is configured to output the optimization result; wherein the establishment process of the simulation model includes: determining the topological structure of the integrated energy system, the topological structure including the equipment of the integrated energy system and the connection attributes between the equipment; Determine a general model of the device and a connector model corresponding to the connection attribute; connect the general model via the connector model to form a simulation model of the integrated energy system; and train the simulation model.
  • FIG. 13 is a structural diagram of an optimization device for an integrated energy system with a processor-memory architecture according to an embodiment of the present invention.
  • the optimization device 1300 of the integrated energy system includes a processor 1301 and a memory 1302; the memory 1302 stores an application program that can be executed by the processor 1301, and is used to make the processor 1301 execute any of the above Optimization methods for integrated energy systems.
  • the optimization methods and optimization devices described in Figs. 11-13 can be used to realize the optimal scheduling of production in the integrated energy system and the peak-shift energy consumption of the factory.
  • the optimization algorithm automatically optimizes the scheduling of controllable resources through the lower-level control system, and accurately reflects the effects of uncontrollable resources. At the same time, it can automatically identify system changes, thereby realizing automatic and flexible changes.
  • the performance of system components can be accurately predicted under various boundary conditions.
  • Various boundary conditions including dynamic price system, accurate weather forecast, etc., can achieve dynamic optimization to 15 minutes Interval input (manual page input or EXCEL import, or automatic import when connected with other information systems).
  • the load demand within the optimized time range needs to be accurately predicted.
  • various methods can be used to import any production plan to ensure the accuracy of load forecasting.
  • pre-determining some optimized operation strategies for selection at any time such as: peak shaving and valley filling, frequency modulation and voltage regulation, and factory peak-shift operation.
  • FIG. 15 is a system structure diagram of an optimized example of the integrated energy system according to the embodiment of the present invention.
  • the gas turbine supplies power to load 1 and load 2 at the same time, and the photovoltaic system and the grid jointly supply power to load 3.
  • the variables marked in Figure 15 represent the power provided or consumed by different devices.
  • the output of the gas turbine is X out,1 ;
  • the input of load 1 is X in,1 ;
  • the input of load 2 is X in,2 ;
  • the output of the power grid is X out,2 ;
  • the output of the photovoltaic system is X out,3 ;
  • the input of the energy storage battery model is X in,4 ;
  • the output of the energy storage battery model is X out,4 ;
  • the input of load 3 is X in,3 and X out,4 .
  • each device also satisfies its own constraint relationship, for example:
  • P GT,max represents the maximum output power of the gas turbine
  • P BT,max represents the maximum battery charge and discharge power
  • C BT represents the battery capacity
  • SOC represents the current storage percentage of the battery.
  • the cost calculation function consumed by the system is:
  • Cost i aX out,1 +bX out,2 +cX in,4 +dX out,4 (20)
  • a, b, c, d are the cost systems, for example: a reflects the operating cost of gas turbine power generation, b reflects the power purchase cost of the grid, and c and d reflect the cost of battery charging and discharging respectively. Therefore, in an optimization period, the total cost of the system is:
  • the optimized variables are gas turbine power generation X out,1 , grid power purchase curve X out,2 , energy storage battery charging and discharging curves X in,4 and X out,4 . That is to determine the optimal power generation, power purchase, charging and discharging strategies according to the load power demand and photovoltaic power supply capacity.
  • a hardware module may include specially designed permanent circuits or logic devices (such as dedicated processors, such as FPGAs or ASICs) to complete specific operations.
  • the hardware module may also include programmable logic devices or circuits temporarily configured by software (for example, including general-purpose processors or other programmable processors) for performing specific operations.
  • programmable logic devices or circuits temporarily configured by software for example, including general-purpose processors or other programmable processors
  • it can be determined according to cost and time considerations.
  • the present invention also provides a machine-readable storage medium that stores instructions for making a machine execute the method described herein.
  • a system or device equipped with a storage medium may be provided, and the software program code for realizing the function of any one of the above embodiments is stored on the storage medium, and the computer (or CPU or MPU of the system or device) ) Read and execute the program code stored in the storage medium.
  • an operating system or the like operating on the computer can also be used to complete part or all of the actual operations through instructions based on the program code.
  • Implementations of storage media used to provide program codes include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), Magnetic tape, non-volatile memory card and ROM.
  • the program code can be downloaded from a server computer or cloud via a communication network.

Abstract

一种综合能源系统的仿真方法、装置和计算机可读存储介质。接收仿真任务；基于所述仿真任务和所述综合能源系统的仿真模型，建立非线性方程组；基于线性规划算法求解非线性方程组以获取仿真结果；输出所述仿真结果；其中所述仿真模型的建立过程包括：确定综合能源系统的拓扑结构，所述拓扑结构包含综合能源系统的设备及所述设备之间的连接属性；确定所述设备的通用模型及对应于所述链接属性的连接件模型；经由所述连接件模型连接所述通用模型，以组成所述综合能源系统的仿真模型；训练所述仿真模型。可以执行各种仿真任务，以获取某一时刻系统的状态，或加入某些操作假设分析出的状态。

Description

综合能源系统的仿真方法、装置和计算机可读存储介质 技术领域
本发明涉及综合能源技术领域，特别是涉及一种综合能源系统的仿真方法、装置和计算机可读存储介质。
背景技术
综合能源服务是指一定区域内利用先进的物理信息技术和创新管理模式，整合区域内煤炭、石油、天然气、电能、热能等多种能源，实现多种异质能源子系统之间的协调规划、优化运行，协同管理、交互响应和互补互济。
可以提供综合能源服务的综合能源系统指在规划、建设和运行等过程中，通过对能源的产生、传输与分配(能源网络)、转换、存储、消费等环节进行有机协调与优化后，形成的能源产供销一体化系统。它主要由供能网络(如供电、供气、供冷/热等网络)、能源交换环节(如CCHP机组、发电机组、锅炉、空调、热泵等)、能源存储环节(储电、储气、储热、储冷等)、终端综合能源供用单元(如微网)和大量终端用户共同构成。
目前，针对综合能源系统的控制过程繁琐，复杂度高。
发明内容
本发明实施方式提出一种综合能源系统的控制方法、装置和计算机可读存储介质。
本发明实施方式的技术方案如下：
一种综合能源系统的仿真方法，所述综合能源系统包含非线性设备，该方法包括：
接收仿真任务；
基于所述仿真任务和所述综合能源系统的仿真模型，建立非线性方程组；
基于线性规划算法求解非线性方程组以获取仿真结果；
输出所述仿真结果；
其中所述仿真模型的建立过程包括：确定综合能源系统的拓扑结构，所述拓扑结构包含综合能源系统的设备及所述设备之间的连接属性；确定所述设备的通用模型及对应于所述连接属性的连接件模型；经由所述连接件模型连接所述通用模型，以组成所述综合能源系统的仿真模型；训练所述仿真模型。
在一个实施方式中，所述综合能源系统还包含线性设备；
该方法还包括预先生成每种线性设备的通用模型和预先生成每种非线性设备的通用模型的过程，其中 生成每种非线性设备的通用模型的过程包括：
针对每种非线性设备的每个目标非线性基理过程，确定其完整的设计点数据；
采用相似性准则支持的相似数与基于设计点数据的相似数的比值，建立所述非线性基理过程的描述公式，得到所述非线性基理过程的通用模型；所述非线性基理过程的通用模型中包括随实际工况参数变更而非线性变化的可变参数；
构建所述实际工况参数与所述可变参数之间的机器学习算法，并建立所述机器学习算法与所述非线性基理过程的通用模型之间的关联关系；
将每种非线性设备的所有目标非线性通用过程的通用模型及其所关联的机器学习算法，构成该种非线性设备的通用模型。
在一个实施方式中，所述训练仿真模型包括：
获取所述仿真模型的运行过程中所述设备的历史数据；
基于所述设备的历史数据训练所述通用模型。
在一个实施方式中，所述基于设备的历史数据训练通用模型包括：基于非线性设备的历史数据，训练非线性设备的通用模型的过程；
其中该过程包括：
针对非线性设备的每个目标非线性基理过程，获取所述非线性设备的所述目标非线性基理过程对应的实际工况参数和可变参数的历史数据，利用所述历史数据对所述机器学习算法进行训练，得到所述目标非线性基理过程的可变参数训练模型；
将所述目标非线性基理过程的可变参数训练模型代入所述目标非线性基理过程的通用模型中，得到所述非线性设备的所述目标非线性基理过程的训练后模型；
将所述非线性设备的所有目标非线性基理过程的训练后模型，构成所述非线性设备的训练后模型。
在一个实施方式中，所述非线性设备包括：燃气轮机、热泵：
所述燃气轮机的目标非线性基理过程包括：膨胀透平中的流量与压力的相关过程、热能与机械能能量转换的过程；
所述热泵的目标非线性基理过程包括：传热过程、热能转换为动能的过程、管道阻力过程、流量与压力的相关过程。
在一个实施方式中，
所述仿真任务包含下列中的至少一个：
关于设备性能监控的仿真任务；包含操作假设条件的仿真任务；关于连接件模型性能监控的仿真任务；关于综合能源系统的整体性能监控的仿真任务。
在一个实施方式中，
所述线性规划算法包括下列中的至少一个：
MIP算法；MILP算法。
一种综合能源系统的仿真装置，所述综合能源系统包含非线性设备，该装置包括：
接收模块，用于接收仿真任务；
方程组建立模块，用于基于所述仿真任务和所述综合能源系统的仿真模型，建立非线性方程组；
求解模块，用于基于线性规划算法求解非线性方程组以获取仿真结果；
输出模块，用于输出所述仿真结果；
其中所述仿真模型的建立过程包括：确定综合能源系统的拓扑结构，所述拓扑结构包含综合能源系统的设备及所述设备之间的连接属性；确定所述设备的通用模型及对应于所述连接属性的连接件模型；经由所述连接件模型连接所述通用模型，以组成所述综合能源系统的仿真模型；训练所述仿真模型。
在一个实施方式中，所述综合能源系统还包含线性设备；
其中所述仿真模型的建立过程还包括预先生成每种线性设备和非线性设备的通用模型的过程，其中生成每种非线性设备的通用模型的过程包括：针对每种非线性设备的每个目标非线性基理过程，确定其完整的设计点数据；采用相似性准则支持的相似数与基于设计点数据的相似数的比值，建立所述非线性基理过程的描述公式，得到所述非线性基理过程的通用模型；所述非线性基理过程的通用模型中包括随实际工况参数变更而非线性变化的可变参数；构建所述实际工况参数与所述可变参数之间的机器学习算法，并建立所述机器学习算法与所述非线性基理过程的通用模型之间的关联关系；将每种非线性设备的所有目标非线性通用过程的通用模型及其所关联的机器学习算法，构成该种非线性设备的通用模型。
在一个实施方式中，所述仿真任务包含下列中的至少一个：
关于设备性能监控的仿真任务；包含操作假设条件的仿真任务；关于连接件模型性能监控的仿真任务；关于综合能源系统的整体性能监控的仿真任务。
在一个实施方式中，所述线性规划算法包括下列中的至少一个：
MIP算法；MILP算法。
一种综合能源系统的仿真装置，包括处理器和存储器；
所述存储器中存储有可被所述处理器执行的应用程序，用于使得所述处理器执行如上任一项所述的综合能源系统的仿真方法。
一种计算机可读存储介质，其中存储有计算机可读指令，该计算机可读指令用于执行如上任一项所述的综合能源系统的仿真方法。
可见，本发明实施方式中，接收仿真任务；基于仿真任务和所述综合能源系统的仿真模型，建立非线性方程组；基于线性规划算法求解非线性方程组以获取仿真结果；输出仿真结果；其中仿真模型的建立过程包括：确定综合能源系统的拓扑结构，所述拓扑结构包含综合能源系统的设备及所述设备之间的连接属 性；确定所述设备的通用模型及对应于所述连接属性的连接件模型；经由所述连接件模型连接所述通用模型，以组成所述综合能源系统的仿真模型；训练所述仿真模型。基于设备的通用模型及连接件模型组成综合能源系统的仿真模型；训练仿真模型，并基于训练后的仿真模型生成综合能源系统的控制命令。因此，通过可复用的通用模型，实现了专业知识的各行业、各用户的共享，便于构建综合能源系统的仿真系统，并提高了综合能源系统的控制效率。
而且，考虑到综合能源系统的实际运行过程中，设备老化、积灰等自然因素或操作人员修改了设备底层参数时，常常会导致内嵌的设备模型偏离实际情况和计算结果存在偏差，本发明实施方式根据实际运转情况以训练方式修正设备的模型参数，从而达到更高的准确度。
针对每种非线性设备的每个目标非线性基理过程，采用相似性准则支持的相似数与基于设计点数据的相似数的比值，建立非线性基理过程的描述公式，得到非线性基理过程的通用模型，使得该模型可适用于一类设备。此外，通用模型中包括随实际工况参数变更而非线性变化的可变参数，并且通过构建实际工况参数与所述可变参数之间的机器学习算法，使得该可变参数可通过机器学习获得，进而使得该通过模型具有自学习能力。还有，针对该种设备的一个具体设备，通过对其每个目标非线性基理过程，获取具体设备的目标非线性基理过程对应的实际工况参数和可变参数的历史数据，并利用历史数据对机器学习算法进行训练，得到目标非线性基理过程的可变参数训练模型，并将目标非线性基理过程的可变参数训练模型代入目标非线性基理过程的通用模型中，可得到具体设备的目标非线性基理过程的训练后模型，也即符合具体设备特性的实例化模型。再有，通过为可变参数预先设置默认值，可在现场不具备训练可变参数的情况下，例如没有足够的历史数据等情况下，可以使得该非线性模型可用。
附图说明
图1为本发明实施方式的综合能源系统的控制方法的流程图。
图2为本发明实施方式基于连接件模型和设备的通用模型组成综合能源系统的仿真模型的示意图。
图3为本发明实施方式设置设备的通用模型的参数的示意图。
图4为本发明实施方式生成非线性设备的通用模型的方法流程图。
图5为本发明实施方式的综合能源系统的控制装置的模块图。
图6为本发明实施方式具有处理器-存储器架构的综合能源系统的控制装置的结构图。
图7为本发明实施方式具综合能源系统的示范性拓扑结构图。
图8为本发明实施方式综合能源系统的仿真方法流程图。
图9为本发明实施方式综合能源系统的仿真装置结构图。
图10为本发明实施方式具有处理器-存储器架构的综合能源系统的仿真装置结构图。
图11为本发明实施方式综合能源系统的优化方法流程图。
图12为本发明实施方式综合能源系统的优化装置结构图。
图13为本发明实施方式具有处理器-存储器架构的综合能源系统的优化装置的结构图。
图14为本发明实施方式综合能源系统的仿真示例的系统结构图。
图15为本发明实施方式综合能源系统的优化示例的系统结构图。
其中，附图标记如下：
Figure PCTCN2019109673-appb-000001
Figure PCTCN2019109673-appb-000002
Figure PCTCN2019109673-appb-000003
具体实施方式
为了使本发明的技术方案及优点更加清楚明白，以下结合附图及实施方式，对本发明进行进一步详细说明。应当理解，此处所描述的具体实施方式仅仅用以阐述性说明本发明，并不用于限定本发明的保护范围。
为了描述上的简洁和直观，下文通过描述若干代表性的实施方式来对本发明的方案进行阐述。实施方式中大量的细节仅用于帮助理解本发明的方案。但是很明显，本发明的技术方案实现时可以不局限于这些细节。为了避免不必要地模糊了本发明的方案，一些实施方式没有进行细致地描述，而是仅给出了框架。下文中，“包括”是指“包括但不限于”，“根据……”是指“至少根据……，但不限于仅根据……”。由于汉语的语言习惯，下文中没有特别指出一个成分的数量时，意味着该成分可以是一个也可以是多个，或可 理解为至少一个。
在本发明实施方式中，利用可重复使用的设备通用模型组成综合能源系统的仿真模型，基于设备的历史数据训练设备的通用模型，从而实现对仿真模型的训练，并由训练后的仿真模型生成综合能源系统的控制命令，从而降低控制复杂度。
图1为本发明实施方式的综合能源系统的控制方法的流程图。
如图1所示，该方法包括：
步骤101：确定综合能源系统的拓扑结构，拓扑结构包含综合能源系统的设备及设备之间的连接属性。
在这里，基于园区的各种能源边界条件(比如，空调机组的运行时间；环境温度和湿度；车间设备的开工情况；车间门窗的开启状况，等等)以及设计目标等输入，确定综合能源系统的拓扑结构。其中，拓扑结构包含综合能源系统的设备及各个设备之间的连接属性。
优选地，综合能源系统中所包含的设备可以包括：吸收制冷(AC)设备、多效蒸发(MED)设备、压缩制冷(CC)设备、多级闪蒸精馏(MSFD)设备、联合热和冷供给(CHC)设备、太阳光伏(PV)设备、冷水储能(CWS)设备、冷热可转换热泵(rHP)设备、电锅炉(EB)设备、反渗透(RO)设备、电化学储能(ECES)设备、汽轮机(ST)设备、燃气锅炉(GB)设备、储水(WS)设备、燃气轮机(GT)设备、风电(WT)设备、燃气轮机进口冷却器(HEX)设备、内燃机(ICE)设备、热泵(HP)设备、冰储-整体化脱盐(isiD)设备、高温储热/低温储热(HTS/LTS)设备、低温储热(ITES)设备、冰热力储能(ITES)设备，等等。
优选地，设备之间的连接属性可以包括：
(1)、设备之间的连接关系，比如A设备需要连接B设备，B设备需要连接C设备；
(2)、设备之间的连接特征，比如当确定设备之间的连接关系后，连接特征具体包括管道连接、电连接、经济学连接，等等。
图7为本发明实施方式综合能源系统的示范性拓扑结构图。
由图7可见，该综合能源系统的拓扑结构包含能源701、能量转换部分702、能量消耗部分703和末端用户处的能量形式704。其中：
能源701包含：天然气70、PV/WIND71和太阳热能72。
能量转换部分702包含：高温热储存73、压缩热泵74、吸收热泵75、燃气轮机76、第一燃气锅炉77、第二燃气锅炉78、第一蒸汽发生器79、第二蒸汽发生器80、背压(back press)轮机81、中温热储存82、压缩空气调节83、吸收调节84、LP锅炉85、冷储存86和低温热储存87。
能量消耗部分703包含：过程功率88、过程中的流89、水处理90、冷水岛91、冷却过程92、压缩空气93和热处理/铸造/焊接94。
末端用户处的能量形式包含：电网95、蒸汽96、制冷97和加热98。
而且，在该拓扑结构中，设备之间具有连接属性，比如高温热储存73和第二蒸汽发生器80之间的管道连接，或过程功率88和电网95之间的电连接，等等。
步骤102：确定设备的通用模型及对应于连接属性的连接件模型。
拓扑结构中的设备通常包含线性设备和非线性设备，该方法还包括预先生成每种线性设备的通用模型和每种非线性设备的通用模型的过程。线性设备的建模过程可以采用一般线性或多元回归等方式，本发明不再赘述。
优选地，在本发明实施方式中，生成每种非线性设备的通用模型的过程包括：
针对每种非线性设备的每个目标非线性基理过程，确定其完整的设计点数据；采用相似性准则支持的相似数与基于设计点数据的相似数的比值，建立所述非线性基理过程的描述公式，得到所述非线性基理过程的通用模型；所述非线性基理过程的通用模型中包括随实际工况参数变更而非线性变化的可变参数；构建所述实际工况参数与所述可变参数之间的机器学习算法，并建立所述机器学习算法与所述非线性基理过程的通用模型之间的关联关系；将每种非线性设备的所有目标非线性通用过程的通用模型及其所关联的机器学习算法，构成该种非线性设备的通用模型。
具体地，连接件模型可以实施为电线模型、管道连接线模型、经济学连接线模型，等等。
步骤103：经由连接件模型连接通用模型，以组成综合能源系统的仿真模型。
比如，可以通过图形界面拖拽等形式经由连接件模型连接通用模型，以组成综合能源系统的仿真模型。其中，针对具有额外输入参数的通用模型，还可以通过可视化箭头形式，标示出额外输入参数。
步骤104：训练仿真模型。
具体的，训练仿真模型包括：获取仿真模型的运行过程中，组成综合能源系统中的各个设备的历史数据；基于这些设备的历史数据训练通用模型。其中，当设备包含线性设备和非线性设备时，基于仿真模型的运行过程中，线性设备的历史数据训练线性设备的通用模型，基于非线性设备的历史数据训练非线性设备的通用模型。
优选地，用户可以自由定义需要训练的对象目标及相应的决定参数，并自由选用训练算法、相应的历史数据以及合适的技术参数。模型的参数可以利用手动录入输入，也可以通过自动优化，根据存储的长期历史数据以及实时数据进行优化，得到准确度更高的模型。训练得到的数据模型以及机理模型是后续预测和优化的数据基础。
优选地，训练非线性设备的通用模型的过程包括：针对非线性设备的每个目标非线性基理过程，获取所述非线性设备的所述目标非线性基理过程对应的实际工况参数和可变参数的历史数据，利用所述历史数据对所述机器学习算法进行训练，得到所述目标非线性基理过程的可变参数训练模型；将所述目标非线性基理过程的可变参数训练模型代入所述目标非线性基理过程的通用模型中，得到所述非线性设备的所述目标非线性基理过程的训练后模型；将所述非线性设备的所有目标非线性基理过程的训练后模型，构成所述 非线性设备的训练后模型。
在一个实施方式中，非线性设备包括：燃气轮机、热泵：所述燃气轮机的目标非线性基理过程包括：膨胀透平中的流量与压力的相关过程、热能与机械能能量转换的过程。
在一个实施方式中，所述热泵的目标非线性基理过程包括：传热过程、热能转换为动能的过程、管道阻力过程、流量与压力的相关过程。
以上示范性描述了综合能源系统的拓扑结构、设备建模和非线性设备的典型实例，本领域技术人员可以意识到，这种描述仅是示范性的，并不用于限定本发明实施方式的保护范围。
步骤105：基于训练后的所述仿真模型生成所述综合能源系统的控制命令。
在这里，可以基于训练后的所述仿真模型生成所述综合能源系统的控制命令。控制命令可以为针对综合能源系统中单个设备的控制命令，也可以是针对综合能源系统的综合控制指令。而且，控制命令可以实施为仿真命令或优化命令。比如，接收包含优化目标的优化任务；将优化目标输入所述仿真模型；使能所述仿真模型输出符合仿真模型内的约束条件且达到所述优化目标的控制命令。
优选地，该方法还包括：接收仿真任务；基于所述仿真任务运行所述仿真模型，并输出仿真结果；其中所述仿真任务包含下列中的至少一个：关于设备性能监控的仿真任务；包含操作假设条件的仿真任务；关于连接件模型性能监控的仿真任务；关于综合能源系统的整体性能监控的仿真任务，等等。
其中：仿真任务具体包括综合能源系统的性能监测、节能管理、系统参数优化运行。优化任务包括：在某段时间内满足某段负荷的最优化调度。
可以将图1所示流程集成到一个软件平台中，基于该软件平台实施图1所示流程。
比如，该软件平台可以遵循通用IOT平台架构，提供微服务，搭建PAAS层及SAAS层。其中，前端通过与本地数据存储层的SCADA相连，或通过OPC UA与其它系统相连获取数据，平台系统具有解析MQTT物联网协议的功能，各现场可通过把不同通讯协议转换为MQTT协议后上发平台，进行数据通讯。遵循IEC61970及IEC61850标准数据信息模型标准，从而不仅实现数据本身的传递，还包括各种各类数据的层次关系、系统连接、业务逻辑，等语义关系的自由建议或接收。
关于软件平台的软件架构：综合能源服务的多种算法功能是紧密相关的，且可以独立运行，彼此交互，协同工作，完成多种多样的能源管理任务，例如：综合能源相关部件模块、大数据分析模块、系统模块、优化模块、仿真模块等，都承载于容器技术之上，支持分布式布置，支持跨平台，支持云端平台。IEMS采用了MYSQL数据库，利用RABBITMQ形成消息、数据配送总线技术，用于算法之间及算法与前端的消息传递、数据传送及任务分发。采用JASON格式进行数据传递。采用CELERY用于DOCKER之间的消息通讯，业务协调。
关于该软件平台的平台架构：主要分成前端数据获取的ADAPTOR层，PaaS层、SaaS层。ADAPTOR层是与用于与SIEMENS SCADA系统或SIEMENS DEOP进行通讯，或与其它遵循OPC UA协议的系统、 设备进行通讯，或直接接收MQTT协议。PaaS层服务中包括了与综合能源管控系统功能相关的服务：数据处理服务、部件模型库服务、模型训练服务、能源系统构建服务、预测服务、优化服务、报表服务、及仪表盘用户界面服务、数据分析工具服务；与信息及通讯相关的通用PaaS层服务包括：消息接收及配送服务，负荷均衡服务、数据管理服务等。SaaS服务实现了：能源网络数据及性能监测及诊断、负荷及发电短期预测、系统参数运行优化、能源生产供给的优化调度管理、工厂错峰用能管理、节能管理、多网络协调管理、预测性维护管理。
另外，该软件平台可以包括仿真环境及优化环境，仿真环境是指某一时刻，系统的状态，可以连接下面的测量点分析出的状态，也可以是加入某些操作假设分析出的状态。与仿真不同，运行优化是指在某一段时间的总体运行最优，是一个时段的问题。它们有各字的仿真求解器及优化求解器，有各自的中间解释器，共用一个场景及问题建立用户图形界面作为设备、测点、位置、连接关系等语义的输入系统，利用各自的解释器形成各自的算法接收的数学问题，进行求解。
因此，本发明实施方式通过该软件平台的设计，将园区的各种能源的边界条件以及设计目标等输入，通过数据处理、模型训练等形成可以重复使用的标准模型，同时产生与历史负荷匹配的机组选型，形成符合最优目标的综合能源系统的管控软件平台。而且，通过系统的容量匹配优化可以减少电、热、气的单位价格，降低各能源模块的设备成本，提高可再生能源的消纳能力，可有效提高系统经济性。另外，在该软件平台建设完成后，将历史数据与实时的最新的数据相融合，进行相应模型修正，再利用各种先进算法，形成实时运行的调度策略，保证系统运行期间始终能够达到设定的最优目标。为后续实时电力市场交易、需求响应，以及其他能源市场实时交易等提供可靠的信息技术基础。
优选的，软件平台还能够实现工业能源数据的可视化，数据的透明化展示。利用大数据分析策略，基于机理模型及人工智能技术，针对用户负荷进行实时预测，实时动态优化发电预测的短期运行，以应对目前市场削峰填谷，阶梯电价，调频调压，工厂节能降耗及未来自由市场交易(期货及现货交易，前一天及当天交易市场)的需求。从中长期来看，软件平台能够进行优化运行管理。针对系统内各部件的中期性能、状态进行监测，并进行实时的能效分析。利用实时收集的各项能源数据，进行仿真分析，制定技改方案规划，以及技术经济性分析。还有，软件平台可充分利用可再生能源、化石燃料、余温余压、新能源等多种资源形式，使之相互配合，通过网、储的灵活运行，建立创新的商业机制、采用智能的手段实现高质量、高效率和经济的区域电、热、冷、气等多种负荷的综合供给，满足终端负荷随机性变动要求。
在本发明实施方式中，优化算法通过下层控制系统自动对可控的资源进行优化调度，对不可控的资源则准确反应其产生的作用。同时系统能够自动识别系统的变化，从而实现自动灵活的改变。平台可以选择优化目标，包括能源成本、二氧化碳排放，效率等。
图2为本发明实施方式基于连接件模型和设备的通用模型组成综合能源系统的仿真模型的示意图。
燃气轮机通用模型202、溴化锂供热机通用模型213、溴化锂制冷机通用模型214、供电负荷通用模型 215、供热负荷通用模型216和供冷负荷通用模型217都已经预先建立。然后，通过拖拽形式，将燃气轮机通用模型202、溴化锂供热机通用模型213、溴化锂制冷机通用模型214相互连接以形成如图2所示的仿真模型。连接件模型可以包含箭头和连接。基于箭头可以为通用模型提供输入；基于连线可以为通用模型提供电学连接、管道连接或经济学连接。比如，箭头201、箭头210、箭头205、箭头206、箭头207、箭头208和箭头209，可以分别为箭头指向的通用模型提供输入。
当基于图2组成综合能源系统的仿真模型后，可以为该仿真模型输入本地化的参数。
图3为本发明实施方式设置设备的通用模型的参数的示意图。
由图3可见，通过触发(比如单击或双击)图2中的燃气轮机通用模型202，进一步展示参数设置区220。在参数设置区220中，展示与燃气轮机通用模型202相关的参数输入框，比如展示展示温度输入框221、压比输入框222、压力恢复输入框223、进口温度输入框224和空气流量输入框225。然后，用户可以在这些参数输入框中分别输入参数，这些参数被赋值到燃气轮机通用模型202中。
本发明实施方式还提出了一种生成非线性设备的通用模型的方法。
图4为本发明实施方式生成非线性设备的通用模型的方法流程图。
如图4所示，该方法包括：
步骤401：针对非线性设备的每个目标非线性基理过程，获取非线性设备的所述目标非线性基理过程对应的实际工况参数和可变参数的历史数据。
本步骤中，基理过程有时也可称为物理过程，如传热过程、电能转换过程、以及前面提到的流量与压力的相关过程等。针对每种非线性设备，可以确定感兴趣的基理过程，即需要建模的基理过程，将这些需要建模的基理过程称为目标基理过程，其中非线性的目标基理过程便可称为目标非线性基理过程。例如，针对燃气轮机，其目标非线性基理过程可包括：膨胀透平中的流量与压力的相关过程、热能与机械能能量转换的过程等；针对热泵，其目标非线性基理过程可包括：传热过程、热能转换为动能的过程、管道阻力过程、流量与压力的相关过程等。
针对每个非线性的基理过程，可根据公开发表的设计参数及厂家提供给用户的普通设计点信息等，复原完整的设计点数据。例如，针对流量与压力的相关过程这一通用模型，其设计点数据可包括压比、空气流量等，根据这些设计点数据可推导其效率、进口阻力以及抽气量等使用者不可得的相关设计参数。
步骤402：采用相似性准则支持的相似数与基于设计点数据的相似数的比值，建立所述非线性基理过程的描述公式，得到所述非线性基理过程的通用模型；所述通用模型中包括随实际工况参数变更而非线性变化的可变参数。
其中，实际工况参数指的是某一具体设备的随实际工况参数变化而变化的参数。例如，随着长时间的使用发生机械磨损而产生变化的尺寸，或随季节变化而变化的温度，或随不同做功情况变化的相关参数等。所述可变参数可具有预先设置的默认值。
由于针对每种非线性设备可能存在不同的型号，例如，以压缩机为例，可能存在功率为5M、50M、500M等不同功率的压缩机，因此为了建立压缩机的通用模型，需要采用相似性准则支持的相似数来代替具体的参数值。例如，仍以上述的流量与压力的相关过程这一通用模型为例，使用流量及压力、功率的相似准则支持的相似数代替具体参数，例如：流量的相似准则可如下式(1)所示：
Figure PCTCN2019109673-appb-000004
其中，G1为流量，T1为温度，P1为压力，G0为相应设计点流量，T0为相应设计点温度，P0为相应设计点压力。
相应地，得到的流量与压力的相关过程的通用模型可如下式(2)所示：
Figure PCTCN2019109673-appb-000005
其中，f()为函数，系数a和b为随实际工况参数变更而非线性变化的可变参数，实际应用中，也可为可变参数a和b设置一个默认值。IGV为进口可转导叶角度。
步骤403：构建所述实际工况参数与所述可变参数之间的机器学习算法。
本步骤中，可基于神经智能网络或支持向量机等机器学习大数据分析的方法来构建所述实际工况参数与所述可变参数之间的机器学习算法
步骤404：每种设备的所有目标非线性基理过程的通用模型及其所关联的机器学习算法构成该种设备的通用模型。
实际应用中，用户购买通用模型之后，需要搭建自己的综合能源系统的仿真模型，此时，每个通用模型需要与现场的具体设备相关联，因此便需要对该通用模型进行实例化(可称为训练)。相应地，通用模型的训练过程包括：
第一步：针对该种设备的一个具体设备的每个目标非线性基理过程，获取所述具体设备的所述目标非线性基理过程对应的实际工况参数和可变参数的历史数据，利用所述历史数据对所对应的机器学习算法进行训练，得到所述目标非线性基理过程的可变参数训练模型。
本步骤中，具体训练时，将实际工况参数的一组历史数据作为输入样本值，将实际工况参数的所述一组历史数据对应的可变参数的历史数据作为输出样本值，利用大量的输入样本值和对应的输出样本值对所述机器学习算法进行训练，便可得到所述可变参数的自学习模型，也称训练模型。
例如，仍以上述的流量与压力的相关过程为例，则可获取现场的燃气轮机的相关实际工况参数的历史 数据及其对应的可变参数的历史数据，得到输入输出样本集，通过训练后可得到可变参数a和b的训练模型。
第二步：将所述目标非线性基理过程的可变参数训练模型代入所述目标非线性基理过程的通用模型中，得到所述具体设备的所述目标非线性基理过程的训练后模型。该训练后模型为具有学习能力的自学习模型。
本步骤中，可根据所述机器学习算法与所述通用模型的关联关系，将所述目标非线性基理过程的可变参数训练模型代入所述目标非线性基理过程的通用模型中。
例如，仍以上述的流量与压力的相关过程为例，将可变参数a和b的当前训练模型输入上述式(2)中，便可得到现场的燃气轮机的压气机的流量与压力的相关过程的通用模型。
第三步：所述具体设备的所有目标非线性基理过程的训练后模型构成所述具体设备的训练后模型。
实际使用时，该训练后模型的输入参数可包括所有目标非线性基理过程的训练后模型所需的输入参数。
可见，建立综合能源系统的过程中需要建立很多设备的模型，这些设备通常包括很多非线性的物理过程(也称基理过程)，如燃气轮机的压气机中的流量与压力的相关过程，机械能转变为压能的过程等，因此这种设备的模型通常为非线性模型。
以上对本发明实施例中的非线性模型的建模方法进行了详细描述。另外，非线性模型通常建模过程比较复杂，但相应的精度也比较高，但若直接采用该非线性模型运行综合能源系统的相应控制(比如，仿真或优化)，则可能会由于非线性模型的运行速度相对较慢而影响整个综合能源系统的仿真实时性。
因此，在本发明实施方式中，还优选对训练后的非线性模型执行线性化处理，以提高综合能源系统的仿真实时性。
在这里，提出一种非线性模型的线性化处理方法，包括：
第一步：针对每个设备的非线性模型，确定所述模型各个输入参数的取值范围。
例如，有效的功率范围为额定工况的50％到110％的范围，此外，还有当地的环境温度变化的范围，环境压力变化的范围等。
第二步：将每个输入参数的取值范围基于复数个插值点划分为复数个子区间。
本步骤中，可基于均衡准则，将每个输入参数的取值范围基于复数个插值点划分为复数个子区间。其中，插值点的确定可以是针对非线性变化剧烈的区域设置较多的插值点，针对非线性变化缓慢的区域设置较少的插值点。
例如，功率范围插入40个点，环境温度插入20个点，环境压力插入5个点等。
第三步：在每个子区间内均衡确定复数个输入样本值。
本步骤中，可基于均衡准则，在每个子区间内均衡确定复数个输入样本值。例如，功率及环境温度等在实际域中可采用等分方法。
第四步：遍历所述模型各个输入参数的输入样本值组合，利用所述非线性模型得到每个输入样本值组 合对应的输出样本值组合。
例如，对应遍历得到的每组输入样本值，均存在一组对应的输出，例如效率输出，或燃料消耗、排放输出，运行成本输出等。
第五步：利用所有输入样本值组合与其所对应的输出样本值组合生成一个张量表。
例如，如上述已知温度、压力、功率三维度的数值，通过查取张量表，可插值可得到效率的数值。
具体，利用该设备的模型进行仿真时，根据各个输入参数的当前值查找所述张量表，并利用从所述张量表中查找到的对应数据进行插值处理，得到对应的输出值。其中，各个输入参数的当前值可以是真实值或者也可以是假设值。
例如，对于一个设备可以有一张或多张表，例如温度、压力、功率对应于效率的一张表，温度、压力、功率对应于排放的表，或对应于其它任何所需参数的表。其中，插值算法可根据实际情况选用，例如，可以选用线性插值法或非线性插值法等。在一个示例中，可对相临较近的点采用线性插值法，对相距较远的点采用非线性插值法等。
通过三维温度、压力、功率的样条插值等方法可得到需要的温度、压力、功绩对应的其它输出变量的数值。
这种通用方法使用通用程序，无论是任何一种具体模型，例如，热泵、内燃机，换热器等模型都可通过这一方法实现。
优选地，将每个输入参数的取值范围基于复数个插值点划分为复数个子区间为：基于均衡准则，将每个输入参数的取值范围基于复数个插值点划分为复数个子区间。
优选地，所述在每个子区间内均衡确定复数个输入样本值为：基于均衡准则，在每个子区间内均衡确定复数个输入样本值。
优选地，在进行仿真或优化时，根据各个输入参数的当前值对所述张量表进行插值处理，得到对应的输出值。
基于上述描述，本发明实施方式还提出了一种综合能源系统的控制装置。
图5为本发明实施方式的综合能源系统的控制装置的模块图。
如图5所示，控制装置500包括：
拓扑结构确定模块501，用于确定综合能源系统的拓扑结构，所述拓扑结构包含综合能源系统的设备及所述设备之间的连接属性；
模型确定模块502，用于确定所述设备的通用模型及对应于所述连接属性的连接件模型；
仿真模型组成模块503，用于经由所述连接件模型连接所述通用模型，以组成所述综合能源系统的仿真模型；
训练模块504，用于训练所述仿真模型；
控制命令生成模块505，用于基于训练后的所述仿真模型生成所述综合能源系统的控制命令。
在一个实施方式中，所述设备包含线性设备和非线性设备；
模型确定模块502，还用于预先生成每种线性设备和非线性设备的通用模型的过程，其中生成每种非线性设备的通用模型的过程包括：针对每种非线性设备的每个目标非线性基理过程，确定其完整的设计点数据；采用相似性准则支持的相似数与基于设计点数据的相似数的比值，建立所述非线性基理过程的描述公式，得到所述非线性基理过程的通用模型；所述非线性基理过程的通用模型中包括随实际工况参数变更而非线性变化的可变参数；构建所述实际工况参数与所述可变参数之间的机器学习算法，并建立所述机器学习算法与所述非线性基理过程的通用模型之间的关联关系；将每种非线性设备的所有目标非线性通用过程的通用模型及其所关联的机器学习算法，构成该种非线性设备的通用模型。
在一个实施方式中，训练模块504，用于获取所述仿真模型的运行过程中所述设备的历史数据；基于所述设备的历史数据训练所述通用模型。
在一个实施方式中，所述基于设备的历史数据训练通用模型包括：基于非线性设备的历史数据，训练非线性设备的通用模型的过程；
所述训练模块504，用于针对非线性设备的每个目标非线性基理过程，获取所述非线性设备的所述目标非线性基理过程对应的实际工况参数和可变参数的历史数据，利用所述历史数据对所述机器学习算法进行训练，得到所述目标非线性基理过程的可变参数训练模型；将所述目标非线性基理过程的可变参数训练模型代入所述目标非线性基理过程的通用模型中，得到所述非线性设备的所述目标非线性基理过程的训练后模型；将所述非线性设备的所有目标非线性基理过程的训练后模型，构成所述非线性设备的训练后模型。
在一个实施方式中，所述非线性设备包括：燃气轮机、热泵：
所述燃气轮机的目标非线性基理过程包括：膨胀透平中的流量与压力的相关过程、热能与机械能能量转换的过程；或
所述热泵的目标非线性基理过程包括：传热过程、热能转换为动能的过程、管道阻力过程、流量与压力的相关过程。
在一个实施方式中，所述控制命令生成模块505，还用于在所述训练仿真模型之后，接收仿真任务；基于所述仿真任务运行所述仿真模型，并输出仿真结果；其中所述仿真任务包含下列中的至少一个：关于设备性能监控的仿真任务；包含操作假设条件的仿真任务；关于连接件模型性能监控的仿真任务；关于综合能源系统的整体性能监控的仿真任务。
在一个实施方式中，控制命令生成模块505，用于接收包含优化目标和约束条件的优化任务；将所述约束条件和所述优化目标输入所述仿真模型；使能所述仿真模型输出符合所述优化目标的控制命令。
本发明实施方式还提出了一种具有处理器和存储器结构的综合能源系统的控制装置。
图6为本发明实施例处理器-存储器结构的、综合能源系统的控制装置的结构图。
如图6所示，综合能源系统的控制装置600包括处理器601和存储器602；
存储器602中存储有可被处理器601执行的应用程序，用于使得处理器601执行如上任一项所述的综合能源系统的控制方法。
其中，存储器602具体可以实施为电可擦可编程只读存储器(EEPROM)、快闪存储器(Flash memory)、可编程程序只读存储器(PROM)等多种存储介质。处理器601可以实施为包括一或多个中央处理器或一或多个现场可编程门阵列，其中现场可编程门阵列集成一或多个中央处理器核。具体地，中央处理器或中央处理器核可以实施为CPU或MCU。
基于上述描述以建立综合能源系统的仿真模型且训练仿真模型，并对训练后的非线性模型执行线性化处理后，可以方便地使用上述处理后的仿真模型执行各种任务，比如执行仿真任务或优化任务。
在本发明实施方式中，综合能源系统的仿真模型平台主要包括仿真环境及优化环境，仿真环境是指某一时刻、综合能源系统的状态，可以连接下面的测量点分析出的状态，也可以是加入某些操作假设分析出的状态。与仿真不同，运行优化环境是指在某一段时间的总体运行最优，是一个时段的问题。
针对仿真环境及优化环境，可以有各自的仿真求解器及优化求解器，有各自的中间解释器，共用一个场景及问题建立用户图形界面作为设备、测点、位置、连接关系等语义的输入系统，利用各自的解释器形成各自的算法接收的数学问题，进行求解。其中：本发明实施方式还实现了基于综合能源系统的仿真模型的仿真方法及优化计算方法，仿真方法实现综合能源系统的性能监测、节能管理、系统参数优化运行；优化方法实现在某段时间内满足某段负荷的最优化调度。
图8为本发明实施方式综合能源系统的仿真方法流程图。
如图8所示，该方法800包括：
步骤801：接收仿真任务。
比如，从用户处接收：设备性能监控的仿真任务；包含操作假设条件的仿真任务；关于连接件模型性能监控的仿真任务；关于综合能源系统的整体性能监控的仿真任务。等等。
步骤802：基于仿真任务和综合能源系统的仿真模型，建立非线性方程组。
在这里，响应于仿真任务，基于仿真模型中的功率平衡公式和性能约束关系，建立非线性方程组。
步骤803：基于线性规划算法求解非线性方程组以获取仿真结果。
在这里，线性规划算法包括下列中的至少一个：混合整数规划(Mixed Integer Programming,MIP)算法；混合整数线性规划算法(Mixed Integer Linear Programming,MILP)算法，等等。
步骤804：输出所述仿真结果；其中所述仿真模型的建立过程包括：确定综合能源系统的拓扑结构，所述拓扑结构包含综合能源系统的设备及所述设备之间的连接属性；确定所述设备的通用模型及对应于所述连接属性的连接件模型；经由所述连接件模型连接所述通用模型，以组成所述综合能源系统的仿真模型； 训练所述仿真模型。
其中，仿真模型中的非线性模型在被训练且被执行了线性化处理后，具有张量表。在步骤803中基于线性规划算法求解非线性方程组以获取仿真结果时，可根据每个非线性模型的输入参数的当前值查找该非线性模型的张量表，并利用从张量表中查找到的对应数值进行插值处理，得到对应的输出值，即为该非线性模型的输出值，从而实现了线性规划算法求解非线性方程组。
图9为本发明实施方式综合能源系统的仿真装置结构图。
如图9所示，综合能源系统的仿真装置900包括：
接收模块901，用于接收仿真任务；
方程组建立模块902，用于基于所述仿真任务和所述综合能源系统的仿真模型，建立非线性方程组；
求解模块903，用于基于线性规划算法求解非线性方程组以获取仿真结果；
输出模块904，用于输出所述仿真结果；其中所述仿真模型的建立过程包括：确定综合能源系统的拓扑结构，所述拓扑结构包含综合能源系统的设备及所述设备之间的连接属性；确定所述设备的通用模型及对应于所述连接属性的连接件模型；经由所述连接件模型连接所述通用模型，以组成所述综合能源系统的仿真模型；训练所述仿真模型。
图10为本发明实施方式具有处理器-存储器架构的综合能源系统的仿真装置结构图。如图10所示，综合能源系统的仿真装置1000包括处理器1001和存储器1002；存储器1002中存储有可被处理器1001执行的应用程序，用于使得处理器1001执行如上任一项所述的综合能源系统的仿真方法。
下面描述基于本发明实施方式的综合能源系统的示范性仿真过程。
图14为本发明实施方式综合能源系统的仿真示例的系统结构图。在图14示意出的综合能源系统的仿真模型的示范结构中，包括经过训练并执行线性化处理后的燃气轮机模型、储能电池模型、光伏模型和三个负载模型。对于如图14所示的简单系统，燃气轮机同时为负荷1和负荷2供电，光伏系统及电网联合为负荷3供电。图14中标出的变量分别表示不同设备提供或消耗的功率。其中，燃气轮机的输出为X out,1；负载1的输入为X in,1；负载2的输入为X in,2；电网的输出为X out,2；光伏系统的输出为X out,3；储能电池模型的输入为X in,4；储能电池模型的输出为X out,4；负载3的输入为X in,3和X out,4。
对于如图14所示的综合能源系统，可以使用本发明实施方式对各设备及整体网络进行仿真计算。在仿真计算时，需要考虑以下几个方面：
(1)、系统网络内的能量(功率)平衡：
具体包括以下表达式：
X out,1-X in,1-X in,2＝0      (3)
X out,2+X out,3-X in,3-X in,4＝0   (4)
(2)、负荷需求特性关系，表达式为：
X in,1＝L 1     (5)
X in,2＝L 2      (6)
X in,3+X out,4＝L 3     (7)
(3)、光伏发电特性关系：
表达式为：X out,3＝PV(m,d,h,weather,…)    (8)
(4)、系统内各设备的性能约束关系。
对于仿真求解，仅需要得到一个满足平衡关系和拓扑关系的解即可，这时，优化目标可以设为方程右边的本应为零的数为一变量，而所有变量的平方和为最小，而把平衡问题化为优化问题，利用各种优化求解器进行求解。调用MILP算法，即可求解得到当前状态下系统各设备的状态及网络内能量流动参数。并具有以下两方面的功能：(1)、与系统内传感器进行对比和校正。(2)、对于不易部署传感器的设备，可使用计算结果了解其运行状态。
仿真求解功能还可预先了解系统内某设备发生变化(含参数变化和)时带来的影响。
例如，图14中光伏输出断开之后，上述第二项平衡方程变为：X out,2+0·X out,3-X in,3-X in,4＝0；保持其它不变，重复上述求解过程，得到一组新的参数，即可反映出这种变化带来的影响。
图11为本发明实施方式综合能源系统的优化方法流程图。
如图11所示，该方法1100包括：
步骤1101：接收包含优化目标的优化任务。
步骤1102：基于所述优化目和所述综合能源系统的仿真模型，建立非线性方程组。
在这里，响应于优化任务，基于仿真模型的功率平衡公式、性能约束关系和接收到的优化目标，建立非线性方程组
步骤1103：基于线性规划算法求解非线性方程组以获取优化结果。
步骤1104：输出所述优化结果；其中所述仿真模型的建立过程包括：确定综合能源系统的拓扑结构，所述拓扑结构包含综合能源系统的设备及所述设备之间的连接属性；确定所述设备的通用模型及对应于所述连接属性的连接件模型；经由所述连接件模型连接所述通用模型，以组成所述综合能源系统的仿真模型；训练所述仿真模型。
其中，仿真模型中的非线性模型在被训练且被执行了线性化处理后，具有张量表。在步骤1103中基于线性规划算法求解非线性方程组以获取优化结果时，可根据每个非线性模型的输入参数的当前值查找该非线性模型的张量表，并利用从张量表中查找到的对应数值进行插值处理，得到对应的输出值，即为该非线性模型的输出值，从而实现了线性规划算法求解非线性方程组。
图12为本发明实施方式综合能源系统的优化装置结构图。
如图12所示，综合能源系统的优化装置1200包括：
接收模块1201，用于接收包含优化目标的优化任务；
方程组建立模块1202，用于基于所述优化目标和所述综合能源系统的仿真模型，建立非线性方程组；
求解模块1203，用于基于线性规划算法求解非线性方程组以获取优化结果；
输出模块1204，用于输出所述优化结果；其中所述仿真模型的建立过程包括：确定综合能源系统的拓扑结构，所述拓扑结构包含综合能源系统的设备及所述设备之间的连接属性；确定所述设备的通用模型及对应于所述连接属性的连接件模型；经由所述连接件模型连接所述通用模型，以组成所述综合能源系统的仿真模型；训练所述仿真模型。
图13为本发明实施方式具有处理器-存储器架构的综合能源系统的优化装置结构图。如图13所示，综合能源系统的优化装置1300包括处理器1301和存储器1302；存储器1302中存储有可被处理器1301执行的应用程序，用于使得处理器1301执行如上任一项所述的综合能源系统的优化方法。
比如，可以利用图11-图13所述的优化方法、优化装置实现综合能源系统中的生产优化调度及工厂错峰用能。优化算法通过下层控制系统自动对可控的资源进行优化调度，对不可控的资源则准确反应其产生的作用，同时能够自动识别系统的变化，从而实现自动灵活的改变。相应地，可以选择优化目标，包括能源成本、二氧化碳排放，效率等。首先，基于准确的性能预测模型，可以实现系统组成部件性能，在各种边界条件下的准确预测，各种边界条件，包括动态价格体系、准确的天气预报等，实现动态以优选为15分种间隔的方式输入(手动页面输入或EXCEL导入，或与其它信息系统相连自动传入)。对优化时间范围内的负荷需求需要进行准确预测。若需要，还可利用各种方式导入任何生产计划的，以保证负荷预测的准确性。同时，还具有预定一些优化运行策略的功能供随时选择，例如：削峰填谷、调频调压，工厂错峰运行等。
下面描述基于本发明实施方式的综合能源系统的示范性优化过程。
图15为本发明实施方式综合能源系统的优化示例的系统结构图。
对于如图15所示的简单系统，燃气轮机同时为负荷1和负荷2供电，光伏系统及电网联合为负荷3供电。图15中标出的变量分别表示不同设备提供或消耗的功率。其中，燃气轮机的输出为X out,1；负载1的输入为X in,1；负载2的输入为X in,2；电网的输出为X out,2；光伏系统的输出为X out,3；储能电池模型的输入为X in,4；储能电池模型的输出为X out,4；负载3的输入为X in,3和X out,4。
(1)、系统的功率平衡方程可以表示为：
X out,1-X in,1-X in,2＝0    (9)
X out,2+X out,3-X in,3-X in,4＝0     (10)
(2)、负荷需求特性关系根据预测结果得到，表示为：
X in,1＝L 1     (11)
X in,2＝L 2     (12)
X in,3+X out,4＝L 3      (13)
(3)、光伏发电特性是时间(比如，月m，日d，小时h)和天气变量(weather)的函数，可以表示为：
X out,3＝PV(m,d,h,weather,…)       (14)
此外，各设备所能提供或消耗的能量还满足各自的约束关系，例如：
X out,1≤P GT,max       (15)
X in,4≤P BT,max      (16)
X out,4≤P BT,max     (17)
X in,4Δt≤C BT(1-SOC)      (18)
X out,4Δt≤C BTSOC     (19)
其中，P GT,max表示燃气轮机最大输出功率，P BT,max表示电池充放电最大功率，C BT表示电池容量，SOC表示电池当前的储电百分比。
考虑平衡关系和约束关系后，对于优化问题，还需要考虑目标函数，在任意时刻时，系统消耗的成本计算函数为：
Cost i＝aX out,1+bX out,2+cX in,4+dX out,4      (20)
其中，a,b,c,d是各项成本的系统，例如：a反映了燃气轮机发电的运行成本，b反映了电网购电成本，c和d分别反映了电池充放电的成本。因此，在一段优化周期之内，系统总消耗的成本Total Cost为：
Figure PCTCN2019109673-appb-000006
将这些方程整合后，即可使用MILP等算法进行求解，得到系统总成本最低的调度方案。其中，被优化的变量为燃气轮机发电量X out,1、电网购电曲线X out,2、储能电池充放电曲线X in,4和X out,4。即根据负荷用电需求、光伏供电能力，来确定最优的发电、购电、充电和放电策略。
需要说明的是，上述各流程和各结构图中不是所有的步骤和模块都是必须的，可以根据实际的需要忽略某些步骤或模块。各步骤的执行顺序不是固定的，可以根据需要进行调整。各模块的划分仅仅是为了便于描述采用的功能上的划分，实际实现时，一个模块可以分由多个模块实现，多个模块的功能也可以由同一个模块实现，这些模块可以位于同一个设备中，也可以位于不同的设备中。
各实施方式中的硬件模块可以以机械方式或电子方式实现。例如，一个硬件模块可以包括专门设计的永久性电路或逻辑器件(如专用处理器，如FPGA或ASIC)用于完成特定的操作。硬件模块也可以包括由软件临时配置的可编程逻辑器件或电路(如包括通用处理器或其它可编程处理器)用于执行特定操作。至于具体采用机械方式，或是采用专用的永久性电路，或是采用临时配置的电路(如由软件进行配置)来实现硬件模块，可以根据成本和时间上的考虑来决定。
本发明还提供了一种机器可读的存储介质，存储用于使一机器执行如本文所述方法的指令。具体地， 可以提供配有存储介质的系统或者装置，在该存储介质上存储着实现上述实施例中任一实施方式的功能的软件程序代码，且使该系统或者装置的计算机(或CPU或MPU)读出并执行存储在存储介质中的程序代码。此外，还可以通过基于程序代码的指令使计算机上操作的操作系统等来完成部分或者全部的实际操作。还可以将从存储介质读出的程序代码写到插入计算机内的扩展板中所设置的存储器中或者写到与计算机相连接的扩展单元中设置的存储器中，随后基于程序代码的指令使安装在扩展板或者扩展单元上的CPU等来执行部分和全部实际操作，从而实现上述实施方式中任一实施方式的功能。
用于提供程序代码的存储介质实施方式包括软盘、硬盘、磁光盘、光盘(如CD-ROM、CD-R、CD-RW、DVD-ROM、DVD-RAM、DVD-RW、DVD+RW)、磁带、非易失性存储卡和ROM。可选择地，可以由通信网络从服务器计算机或云上下载程序代码。
上文通过附图和优选实施例对本发明进行了详细展示和说明，然而本发明不限于这些已揭示的实施例，基与上述多个实施例本领域技术人员可以知晓，可以组合上述不同实施例中的代码审核手段得到本发明更多的实施例，这些实施例也在本发明的保护范围之内。

Claims (13)

  1. 一种综合能源系统的仿真方法(800)，其特征在于，所述综合能源系统包含非线性设备，该方法包括：
    接收仿真任务(801)；
    基于所述仿真任务和所述综合能源系统的仿真模型，建立非线性方程组(802)；
    基于线性规划算法求解非线性方程组以获取仿真结果(803)；
    输出所述仿真结果(804)；
    其中所述仿真模型的建立过程包括：确定综合能源系统的拓扑结构，所述拓扑结构包含综合能源系统的设备及所述设备之间的连接属性；确定所述设备的通用模型及对应于所述连接属性的连接件模型；经由所述连接件模型连接所述通用模型，以组成所述综合能源系统的仿真模型；训练所述仿真模型。
  2. 根据权利要求1所述的综合能源系统的仿真方法(800)，其特征在于，所述综合能源系统还包含线性设备；
    该方法还包括预先生成每种线性设备的通用模型的过程和预先生成每种非线性设备的通用模型的过程(400)，其中生成每种非线性设备的通用模型的过程(400)包括：
    针对每种非线性设备的每个目标非线性基理过程，确定其完整的设计点数据(401)；
    采用相似性准则支持的相似数与基于设计点数据的相似数的比值，建立所述非线性基理过程的描述公式，得到所述非线性基理过程的通用模型；所述非线性基理过程的通用模型中包括随实际工况参数变更而非线性变化的可变参数(402)；
    构建所述实际工况参数与所述可变参数之间的机器学习算法(403)；
    将每种非线性设备的所有目标非线性通用过程的通用模型及其所关联的机器学习算法，构成该种非线性设备的通用模型(404)。
  3. 根据权利要求2所述的综合能源系统的仿真方法(800)，其特征在于，所述训练仿真模型(104)包括：
    获取所述仿真模型的运行过程中所述设备的历史数据；
    基于所述设备的历史数据训练所述通用模型。
  4. 根据权利要求3所述的综合能源系统的仿真方法(800)，其特征在于，
    所述基于设备的历史数据训练通用模型包括：基于非线性设备的历史数据，训练非线性设备的通用模型的过程；
    其中该过程包括：
    针对非线性设备的每个目标非线性基理过程，获取所述非线性设备的所述目标非线性基理过程对应的实际工况参数和可变参数的历史数据，利用所述历史数据对所述机器学习算法进行训练，得到所述目标非线性基理过程的可变参数训练模型；
    将所述目标非线性基理过程的可变参数训练模型代入所述目标非线性基理过程的通用模型中，得到所述非线性设备的所述目标非线性基理过程的训练后模型；
    将所述非线性设备的所有目标非线性基理过程的训练后模型，构成所述非线性设备的训练后模型。
  5. 根据权利要求2所述的综合能源系统的仿真方法(800)，其特征在于，
    所述非线性设备包括：燃气轮机、热泵：
    所述燃气轮机的目标非线性基理过程包括：膨胀透平中的流量与压力的相关过程、热能与机械能能量转换的过程；
    所述热泵的目标非线性基理过程包括：传热过程、热能转换为动能的过程、管道阻力过程、流量与压力的相关过程。
  6. 根据权利要求1-5中任一项所述的综合能源系统的仿真方法(800)，其特征在于，
    所述仿真任务包含下列中的至少一个：
    关于设备性能监控的仿真任务；包含操作假设条件的仿真任务；关于连接件模型性能监控的仿真任务；关于综合能源系统的整体性能监控的仿真任务。
  7. 根据权利要求1-5中任一项所述的综合能源系统的仿真方法(800)，其特征在于，
    所述线性规划算法包括下列中的至少一个：
    混合整数规划MIP算法；混合整数线性规划MILP算法。
  8. 一种综合能源系统的仿真装置(900)，其特征在于，所述综合能源系统包含非线性设备，该装置(900)包括：
    接收模块(901)，用于接收仿真任务；
    方程组建立模块(902)，用于基于所述仿真任务和所述综合能源系统的仿真模型，建立非线性方程组；
    求解模块(903)，用于基于线性规划算法求解非线性方程组以获取仿真结果；
    输出模块(904)，用于输出所述仿真结果；
    其中所述仿真模型的建立过程包括：确定综合能源系统的拓扑结构，所述拓扑结构包含综合能源系统的设备及所述设备之间的连接属性；确定所述设备的通用模型及对应于所述连接属性的连接件模型；经由所述连接件模型连接所述通用模型，以组成所述综合能源系统的仿真模型；训练所述仿真模型。
  9. 根据权利要求8所述的综合能源系统的仿真装置(900)，其特征在于，所述综合能源系统还包含线性设备；
    所述仿真模型的建立过程还包括预先生成每种线性设备的通用模型的过程和预先生成非线性设备的通用模型的过程，其中生成每种非线性设备的通用模型的过程包括：针对每种非线性设备的每个目标非线性基理过程，确定其完整的设计点数据；采用相似性准则支持的相似数与基于设计点数据的相似数的比值，建立所述非线性基理过程的描述公式，得到所述非线性基理过程的通用模型；所述非线性基理过程的通用 模型中包括随实际工况参数变更而非线性变化的可变参数；构建所述实际工况参数与所述可变参数之间的机器学习算法，并建立所述机器学习算法与所述非线性基理过程的通用模型之间的关联关系；将每种非线性设备的所有目标非线性通用过程的通用模型及其所关联的机器学习算法，构成该种非线性设备的通用模型。
  10. 根据权利要求8所述的综合能源系统的仿真装置(900)，其特征在于，
    所述仿真任务包含下列中的至少一个：
    关于设备性能监控的仿真任务；包含操作假设条件的仿真任务；关于连接件模型性能监控的仿真任务；关于综合能源系统的整体性能监控的仿真任务。
  11. 根据权利要求8所述的综合能源系统的仿真装置(900)，其特征在于，
    所述线性规划算法包括下列中的至少一个：
    MIP算法；MILP算法。
  12. 一种综合能源系统的仿真装置(1000)，其特征在于，包括处理器(1001)和存储器(1002)；
    所述存储器(1002)中存储有可被所述处理器(1001)执行的应用程序，用于使得所述处理器(1001)执行如权利要求1至7中任一项所述的综合能源系统的仿真方法(800)。
  13. 一种计算机可读存储介质，其特征在于，其中存储有计算机可读指令，该计算机可读指令用于执行如权利要求1至7中任一项所述的综合能源系统的仿真方法(800)。
PCT/CN2019/109673 2019-09-30 2019-09-30 综合能源系统的仿真方法、装置和计算机可读存储介质 WO2021062753A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
EP19948148.2A EP4024279A4 (en) 2019-09-30 2019-09-30 INTEGRATED ENERGY SYSTEM SIMULATION METHOD, APPARATUS, AND COMPUTER READABLE INFORMATION MEDIA
PCT/CN2019/109673 WO2021062753A1 (zh) 2019-09-30 2019-09-30 综合能源系统的仿真方法、装置和计算机可读存储介质
US17/764,529 US20220358268A1 (en) 2019-09-30 2019-09-30 Method and Apparatus for Simulating Integrated Energy System, and Computer-Readable Storage Medium
CN201980100606.3A CN114503120A (zh) 2019-09-30 2019-09-30 综合能源系统的仿真方法、装置和计算机可读存储介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/109673 WO2021062753A1 (zh) 2019-09-30 2019-09-30 综合能源系统的仿真方法、装置和计算机可读存储介质

Publications (1)

Publication Number Publication Date
WO2021062753A1 true WO2021062753A1 (zh) 2021-04-08

Family

ID=75336715

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/109673 WO2021062753A1 (zh) 2019-09-30 2019-09-30 综合能源系统的仿真方法、装置和计算机可读存储介质

Country Status (4)

Country Link
US (1) US20220358268A1 (zh)
EP (1) EP4024279A4 (zh)
CN (1) CN114503120A (zh)
WO (1) WO2021062753A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023028860A1 (zh) * 2021-08-31 2023-03-09 西门子股份公司 流体网络系统的优化方法及装置
CN117077287A (zh) * 2023-08-16 2023-11-17 小米汽车科技有限公司 车身大型压铸件优化方法和装置

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113932474B (zh) * 2021-11-15 2023-06-06 江苏科技大学 一种热泵多效蒸发耦合式水处理系统及其工作方法
WO2023245403A1 (zh) * 2022-06-21 2023-12-28 西门子股份公司 流体网络系统的优化方法及装置
EP4322076A1 (en) * 2022-08-11 2024-02-14 Vito NV A method and system for characterizing thermal energy distribution in thermal energy exchange systems
WO2024036628A1 (zh) * 2022-08-19 2024-02-22 西门子股份公司 能源系统的拓扑建模方法及装置
WO2024040608A1 (zh) * 2022-08-26 2024-02-29 西门子股份公司 能源管理系统的模型训练方法、装置及存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016061741A1 (en) * 2014-10-21 2016-04-28 Accenture Global Services Limited System, method, and apparatus for capacity determination for micro grid, and tangible computer readable medium
CN105939029A (zh) * 2016-06-21 2016-09-14 中国南方电网有限责任公司电网技术研究中心 综合能源系统规划方案获取方法和系统
CN106447113A (zh) * 2016-10-08 2017-02-22 东南大学 一种基于运行优化模型的多区域综合能源系统运行方法
CN109583141A (zh) * 2018-12-29 2019-04-05 国网江苏省电力有限公司南京供电分公司 一种计及电转气耦合的电-气互联系统机组组合线性模型及系统

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150184549A1 (en) * 2013-12-31 2015-07-02 General Electric Company Methods and systems for enhancing control of power plant generating units

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016061741A1 (en) * 2014-10-21 2016-04-28 Accenture Global Services Limited System, method, and apparatus for capacity determination for micro grid, and tangible computer readable medium
CN105939029A (zh) * 2016-06-21 2016-09-14 中国南方电网有限责任公司电网技术研究中心 综合能源系统规划方案获取方法和系统
CN106447113A (zh) * 2016-10-08 2017-02-22 东南大学 一种基于运行优化模型的多区域综合能源系统运行方法
CN109583141A (zh) * 2018-12-29 2019-04-05 国网江苏省电力有限公司南京供电分公司 一种计及电转气耦合的电-气互联系统机组组合线性模型及系统

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
GE HAILIN ,CHEN HAOYONG ,QIU MING: "Dynamic Simulation of Integrated Regional Energy System Based on Bond Graph", SHAANXI ELECTRIC POWER, vol. 45, no. 2, 20 February 2017 (2017-02-20), pages 37 - 42, XP055798040, ISSN: 1673-7598 *
See also references of EP4024279A4 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023028860A1 (zh) * 2021-08-31 2023-03-09 西门子股份公司 流体网络系统的优化方法及装置
CN117077287A (zh) * 2023-08-16 2023-11-17 小米汽车科技有限公司 车身大型压铸件优化方法和装置

Also Published As

Publication number Publication date
EP4024279A1 (en) 2022-07-06
EP4024279A4 (en) 2023-05-10
US20220358268A1 (en) 2022-11-10
CN114503120A (zh) 2022-05-13

Similar Documents

Publication Publication Date Title
WO2021062748A1 (zh) 综合能源系统的优化方法、装置和计算机可读存储介质
WO2021062753A1 (zh) 综合能源系统的仿真方法、装置和计算机可读存储介质
WO2021062749A1 (zh) 综合能源系统的控制方法、装置和计算机可读存储介质
CN113657031B (zh) 基于数字孪生的供热调度自动化实现方法、系统及平台
CN109858759B (zh) 一种工业园区综合能源平衡调度方法
US9705327B2 (en) Self-learning, real-time, data-driven power metering system
Jennings et al. Modelling and optimization of retrofitting residential energy systems at the urban scale
US20120239164A1 (en) Graphical language for optimization and use
CN113947236B (zh) 综合能源调度方法、计算设备及介质
CN113888132A (zh) 一种大型工业企业的能源管理系统
CN111950122A (zh) 园区综合能源系统运行优化方法
Li et al. Study on operation strategy and load forecasting for distributed energy system based on Chinese supply-side power grid reform
La Bella et al. Data-driven modelling and optimal management of district heating networks
CN113821008A (zh) 一种大型工业企业的能源管理控制方法及能源管理中心
Sawant et al. Experimental demonstration of grid-supportive scheduling of a polygeneration system using economic-MPC
CN103632045B (zh) 一种评估省级电网发电调度理想度的计算方法
CN111105090A (zh) 一种基于智能算法的分布式能源系统优化调度方法及装置
CN115758763A (zh) 一种计及源荷不确定性的多能流系统优化配置方法及系统
Chen et al. Stochastic planning of integrated energy system based on correlation scenario generation method via Copula function considering multiple uncertainties in renewable energy sources and demands
Rafii-Tabrizi Smart Electrical and Thermal Energy Supply for Nearly Zero Energy Buildings
WO2023004809A1 (zh) 一种智慧能源管理系统的建模方法、装置和存储介质
Wu et al. Electricity Market Clearing Model Considering Uncertainty of Wind Power
EP4024259A1 (en) Linearization processing method and device for nonlinear model, and storage medium
Rangstedt Canholm Digital district heating: Developing the heating value proposition with the help of digitalisation
Shendryk et al. Decision-Making Support on Energy Management in District Heating

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19948148

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2019948148

Country of ref document: EP

Effective date: 20220328