CN111476386A - Energy system maintenance method, device, system and storage medium - Google Patents
Energy system maintenance method, device, system and storage medium Download PDFInfo
- Publication number
- CN111476386A CN111476386A CN202010454661.7A CN202010454661A CN111476386A CN 111476386 A CN111476386 A CN 111476386A CN 202010454661 A CN202010454661 A CN 202010454661A CN 111476386 A CN111476386 A CN 111476386A
- Authority
- CN
- China
- Prior art keywords
- model
- digital twin
- energy
- equipment
- twin model
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
Images
Classifications
-
- 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
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06313—Resource planning in a project environment
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- General Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Life Sciences & Earth Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Quality & Reliability (AREA)
- Evolutionary Computation (AREA)
- Operations Research (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Biodiversity & Conservation Biology (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The present disclosure provides a method, an apparatus, a system and a storage medium for maintaining an energy system, wherein the method comprises: the method includes the steps of constructing a digital twin model corresponding to energy equipment in an energy system, obtaining running state information corresponding to the energy equipment by using the digital twin model according to first equipment running data corresponding to the energy equipment, and generating maintenance strategy information corresponding to the energy equipment based on the running state information. The disclosed methods, apparatus, and storage media provide for efficient energy device predictive maintenance; the workload in the equipment maintenance process is reduced, the working efficiency is improved, and the flexibility and the convenience of the energy system maintenance are improved; by training the same type of equipment, model fine tuning is carried out on actual data of respective physical models after the models are deployed, time consumed by model training is reduced, and customized models are realized.
Description
Technical Field
The present invention relates to the field of equipment maintenance technologies, and in particular, to a method, an apparatus, a system, and a storage medium for maintaining an energy system.
Background
Energy equipment in energy systems of factory level, community level, user level and the like needs to be checked regularly or overhauled after a fault problem occurs. At present, the energy equipment is usually maintained in a planned way by adopting a manual verification mode regularly, the maintenance effect mainly depends on the experience of an engineer in charge of maintenance and personal judgment, and misjudgment and missed judgment are easy to occur, so that the maintenance efficiency is low, the equipment maintenance is inaccurate and untimely, and the like. To ensure that the energy device is able to operate efficiently, the field engineer may repair the problem after the fault is discovered, but due to the complexity of the energy system and other limitations, when the fault is discovered, the energy utilization performance of the energy system has fallen below a normal level, while abnormal operation of the energy device may waste a large amount of energy.
Disclosure of Invention
In view of the above, an object of the present invention is to provide a method, an apparatus, a system and a storage medium for maintaining an energy system.
According to one aspect of the present disclosure, there is provided an energy system maintenance method, including: constructing a digital twin model corresponding to an energy device in an energy system; acquiring operating state information corresponding to the energy device using the digital twin model and according to first device operating data corresponding to the energy device; and generating maintenance strategy information corresponding to the energy equipment based on the operation state information.
Optionally, the constructing a digital twin model corresponding to an energy device in an energy system comprises: generating a universal digital twin model corresponding to the energy device type; acquiring an equipment operation data set corresponding to energy equipment belonging to the same energy equipment type as a training sample; training the universal digital twin model based on the training samples; and constructing the digital twin model by using the trained general digital twin model.
Optionally, the constructing the digital twin model by using the trained universal digital twin model includes: setting the universal digital twin model corresponding to the energy equipment according to the type of the energy equipment to which the energy equipment belongs; adjusting the universal digital twin model based on second device operational data corresponding to the energy device to generate the digital twin model.
Optionally, storing all trained universal digital twin models in a master node; the energy equipment type sent by each workstation is obtained, and the main node is controlled to send the universal digital twin model to the corresponding workstation based on the energy equipment type; and when the workstation deploys the universal digital twin model, controlling the workstation to adjust the universal digital twin model by using the second equipment operation data to generate the digital twin model.
Optionally, the workstation is controlled to send the operating state information output by the digital twin model to the master node, and the master node is controlled to generate the maintenance policy information based on the operating state information.
Optionally, the digital twin model is controlled to obtain the first device operation data and the second device operation data collected by a sensor network corresponding to the energy device.
Optionally, the sensor network comprises: a sensor for acquiring the first device operational data and the second device operational data.
Optionally, the training of the universal digital twin model based on the training samples comprises training of the universal digital twin model based on the training samples by using a preset machine learning algorithm, wherein the machine learning algorithm comprises a Meta-learning Meta L earning algorithm.
Optionally, the universal digital twinning model comprises: at least one sub-model; the generating of the universal digital twin model corresponding to the energy device type includes: obtaining modeling information corresponding to the type of the energy equipment; wherein the modeling information comprises: device attributes, device functions, and device feature information; generating the sub-model according to the modeling information; wherein the sub-models comprise: one or more of a device state identification model, a fault model, a device lifetime model, a device operational transient model, a device operational steady state model.
Optionally, the digital twinning model comprises: device information corresponding to an energy device, at least one of the submodels.
Optionally, the obtaining the operation state information corresponding to the energy device according to the device operation data by using the digital twin model comprises: inputting the first equipment operation data into the corresponding digital twin model, and performing simulation operation through the digital twin model to obtain the operation state information; wherein the operating state information includes at least one of historical, current, and predicted operating state information.
Optionally, the generating maintenance policy information corresponding to the energy device based on the operation state information includes: and evaluating the energy equipment based on the running state information, and generating the maintenance strategy information according to the evaluation and processing result.
According to another aspect of the present disclosure, there is provided an energy system maintenance apparatus including: the model building module is used for building a digital twin model corresponding to energy equipment in the energy system; the model operation module is used for acquiring operation state information corresponding to the energy equipment by using the digital twin model according to first equipment operation data corresponding to the energy equipment; and the maintenance strategy making module is used for generating maintenance strategy information corresponding to the energy equipment based on the operation state information.
Optionally, the model building module includes: a general model generation unit for generating a general digital twin model corresponding to the energy device type; acquiring an equipment operation data set corresponding to energy equipment belonging to the same energy equipment type as a training sample; training the universal digital twin model based on the training samples; and the equipment model generating unit is used for constructing the digital twin model by utilizing the trained general digital twin model.
Optionally, the device model generating unit is specifically configured to set the universal digital twin model corresponding to the energy device according to the type of the energy device to which the energy device belongs; adjusting the universal digital twin model based on second device operational data corresponding to the energy device to generate the digital twin model.
Optionally, the model building module includes: the model deployment unit is used for storing all the trained universal digital twin models in the main node; the energy equipment type sent by each workstation is obtained, and the main node is controlled to send the universal digital twin model to the corresponding workstation based on the energy equipment type; and the equipment model generating unit is used for controlling the workstation to adjust the universal digital twin model by using the second equipment operation data to generate the digital twin model when the workstation deploys the universal digital twin model.
Optionally, the maintenance policy making module is further configured to control the workstation to send the operation state information output by the digital twin model to the master node, and control the master node to generate the maintenance policy information based on the operation state information.
Optionally, the model operation module is further configured to control the digital twin model to obtain the first device operation data and the second device operation data acquired by the sensor network corresponding to the energy device.
Optionally, the sensor network comprises: sensor for detecting operating data of the first and second installation
Optionally, the general model generation unit is further configured to train the general digital twin model based on the training samples by using a preset machine learning algorithm, where the machine learning algorithm includes a Meta learning Meta L earning algorithm.
Optionally, the universal digital twinning model comprises: at least one sub-model; the general model generating unit is further used for acquiring modeling information corresponding to the type of the energy equipment; wherein the modeling information comprises: device attributes, device functions, and device feature information; generating the sub-model according to the modeling information; wherein the sub-models comprise: one or more of a device state identification model, a fault model, a device lifetime model, a device operational transient model, a device operational steady state model.
Optionally, the digital twinning model comprises: device information corresponding to an energy device, at least one of the submodels.
Optionally, the model operation module is configured to input the first device operation data into the corresponding digital twin model, and perform simulation operation through the digital twin model to obtain the operation state information; wherein the operating state information includes at least one of historical, current, and predicted operating state information.
Optionally, the maintenance policy making module is configured to evaluate the energy device based on the operation state information, and generate the maintenance policy information according to an evaluation result.
According to yet another aspect of the present disclosure, there is provided an energy system maintenance system including: energy systems and digital twinning systems; the data twinning system comprises: the energy system maintenance device as described above.
According to still another aspect of the present disclosure, there is provided an energy system maintenance apparatus including: a memory; and a processor coupled to the memory, the processor configured to perform the method as described above based on instructions stored in the memory.
According to yet another aspect of the present disclosure, a computer-readable storage medium is provided, which stores computer instructions for execution by a processor to perform the method as described above.
The energy system maintenance method, the device, the system and the storage medium construct a digital twin model corresponding to energy equipment in an energy system, obtain running state information corresponding to the energy equipment by using the digital twin model according to first equipment running data corresponding to the energy equipment, and generate maintenance strategy information corresponding to the energy equipment based on the running state information; the product quality of the energy equipment can be improved, and efficient predictive maintenance of the energy equipment is provided; the workload in the equipment maintenance process is reduced, the working efficiency is improved, and the flexibility and the convenience of energy system maintenance are improved.
Drawings
In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present disclosure, and other drawings can be obtained by those skilled in the art without inventive exercise.
Fig. 1 is a schematic flow diagram of one embodiment of an energy system maintenance method according to the present disclosure;
FIG. 2 is a diagram illustrating the correspondence between energy devices and a digital twin model;
FIG. 3 is a schematic diagram of a first device operating data of an energy device being collected by a sensor and sent to a digital twin model;
FIG. 4 is a schematic flow diagram of constructing a digital twin model in an embodiment of an energy system maintenance method according to the present disclosure;
FIG. 5 is a schematic diagram of training a generic digital twin model;
FIG. 6 is a schematic flow diagram of constructing a digital twin model in an embodiment of an energy system maintenance method according to the present disclosure;
FIG. 7 is a schematic diagram of a digital twinning model deployment;
FIG. 8 is a schematic diagram of a digital twinning model acquisition;
FIG. 9 is a schematic diagram of an application of a digital twinning model;
fig. 10 is a block schematic diagram of one embodiment of an energy system maintenance device according to the present disclosure;
fig. 11 is a block schematic diagram of a model building module in an embodiment of an energy system maintenance device according to the present disclosure;
fig. 12 is a block schematic diagram of one embodiment of an energy system maintenance system according to the present disclosure;
fig. 13 is a block schematic diagram of another embodiment of an energy system maintenance device according to the present disclosure.
Detailed Description
The present disclosure now will be described more fully hereinafter with reference to the accompanying drawings, in which exemplary embodiments of the disclosure are shown. The technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
Fig. 1 is a schematic flow chart of an embodiment of an energy system maintenance method according to the present disclosure, as shown in fig. 1:
The energy equipment is equipment operating in an energy system and comprises power transmission and distribution equipment, motor equipment, heating equipment, ventilation equipment, air conditioning equipment, conventional electrical equipment and the like. The digital twin model is a digital model corresponding to various energy devices in an energy system, is obtained by integrating multidisciplinary, multi-physical quantity and multi-scale mapping of the energy devices to a virtual space, has the size, shape and structure completely the same as those of the energy devices, can complete the same actions and tasks as those of entity devices, and can be various existing digital twin models.
And 102, acquiring running state information corresponding to the energy equipment by using the digital twin model according to the first equipment running data corresponding to the energy equipment.
The operating state of the energy device can be simulated by using the digital twin model and according to the first device operating data, and the operating state information corresponding to the energy device is obtained. The operation state information includes at least one of historical, current and predicted operation state information, and the operation state information specifically includes information of equipment fault, steady-state operation, transient operation, service life, equipment parameters and the like.
The first device operation data is voltage, current, power, pressure, temperature and the like of the energy device. First device operation data collected by a sensor network corresponding to the energy device can be acquired, the sensor network comprises sensors used for collecting the first device operation data, and the sensors can be various.
And 103, generating maintenance strategy information corresponding to the energy equipment based on the operation state information.
The operation data of the first equipment can be input into a corresponding digital twin model, and simulation operation is carried out through the digital twin model so as to obtain operation state information; and evaluating the energy equipment based on the running state information, and generating maintenance strategy information according to the evaluation and processing result. The generated maintenance strategy information comprises information such as maintenance plans, parts needing to be overhauled, overhauling contents and means and the like of the energy equipment, and predictive maintenance of the energy equipment can be realized.
In one embodiment, as shown in FIG. 2, the digital twin model is in a one-to-one correspondence with a remote or nearby energy device. The digital twin model is a digital version of the energy device, and after creation, the digital twin model can acquire historical, current, and predicted operating state information in a digitized form by acquiring first device operating data corresponding to the energy device. Each digital twin model may include an operation and maintenance database configured to store maintenance data corresponding to the energy device and inspection data associated with the energy device. In the process of constructing the digital twin model, the digital twin model needs to be consistent with actual energy equipment, the digital twin model comprises sub models such as a fault model, an asset life model, a transient operation model and a steady operation model, and the types of the sub models are set according to the characteristics of the energy equipment.
In one embodiment, as shown in fig. 3, first device operation data collected by a sensor in a sensor network may be acquired, the sensor is deployed in an energy device, and the sensor may be a temperature sensor, a humidity sensor, an illumination sensor, a pressure sensor, a current sensor, a voltage sensor, a power sensor, or the like. The sensor network transmits first equipment operation data through a CAN bus and the like, and obtains operation state information corresponding to the energy equipment by using a digital twin model according to the first equipment operation data. The operation data of the first equipment collected by the sensor is transmitted to each submodel of the digital twin model, and the operation state information is output through each submodel, so that the variable of the energy equipment which is difficult to obtain through observation can be obtained, and the early failure of the energy equipment can be found more accurately and timely.
In one embodiment, generating the maintenance policy information corresponding to the energy device based on the operation state information may employ various methods. For example, the energy device is evaluated based on the operating state information, and the maintenance policy information is generated according to the evaluation result. The health state or the running state of the energy equipment can be evaluated and judged by analyzing the running state information of the energy equipment in the time dimension, and a corresponding maintenance strategy is formulated for the energy equipment according to the evaluation and judgment result. In the maintenance strategy making, the existing pattern recognition related method, such as Bayes method, Monte Carlo method, deep learning method, etc., can be adopted.
Fig. 4 is a schematic flow chart of constructing a digital twin model in an embodiment of the energy system maintenance method according to the present disclosure, as shown in fig. 4:
The energy equipment in the energy system is classified, the energy equipment type can be a power generation type, an energy storage type, a current transformation type, an electricity utilization type and the like, the generated general digital twin model corresponding to the energy equipment type can be a power generation model, an energy storage model, a current transformation model, an electricity utilization model and the like, and the general digital twin model can be an existing deep learning model and the like.
And 403, training the universal digital twin model based on the training samples.
And step 404, constructing a digital twin model by using the trained general digital twin model.
For example, a preset machine learning algorithm is used and the universal digital twin model is trained based on the training samples, the machine learning algorithm comprises a Meta learning Meta L earning algorithm and the like.
The universal digital twin model comprises at least one sub-model; and obtaining modeling information corresponding to the type of the energy equipment, wherein the modeling information comprises equipment attribute, equipment function and equipment characteristic information, and the equipment characteristic information comprises parameter information of normal operation of the equipment and the like. And generating a sub-model according to the modeling information, wherein the sub-model comprises one or more of an equipment state identification model, a fault model, an equipment service life model, an equipment operation transient model and an equipment operation steady-state model. The digital twin model includes device information corresponding to the energy device, at least one sub model, and the like, the device information including information of a name, a function, and the like of the device.
In one embodiment, machine learning algorithms such as Meta L earning algorithm are used for training each sub-model of the universal digital twin model based on training samples, as shown in FIG. 5, the sub-models of the universal digital twin model comprise a device state identification model, and the device state identification model comprises a data input interface, a deep neural network identifier and an identification result expression interface.
The method comprises the steps of training an equipment state identification model by adopting a Meta L earning algorithm in deep learning, inputting a data set into operation data of equipment in each state, and marking a data label to be an equipment operation state under one-hot coding.
In the training process, a Meta L earning algorithm is adopted, so that the equipment state identification model learns not only the relation between the equipment state and the operation data, but also a method for learning how to learn the operation data set of the type of equipment, in the subsequent model deployment process, the learning aiming at the operation data of the type of equipment corresponding to the model becomes very fast, and the feasibility of the personalized training of the digital twin model in the use of large-scale equipment is improved.
The method for constructing the digital twin model by using the trained general digital twin model can adopt various methods. Fig. 6 is a schematic flow chart of constructing a digital twin model in an embodiment of the energy system maintenance method according to the present disclosure, as shown in fig. 6:
And step 602, acquiring second device operation data corresponding to the energy device, and adjusting the general digital twin model based on the second device operation data to generate a digital twin model.
The second device operation data acquired by the sensors in the sensor network can be acquired, the general digital twin model is adjusted based on the second device operation data, and various existing adjusting methods can be used. For example, a loss function corresponding to the energy equipment is set, and model parameters of the universal digital twin model are adjusted based on the second equipment operation data until the loss value of the loss function meets a preset ending condition.
In the system configuration, assuming that digital twin models of energy devices belonging to the same energy device type are the same, a general digital twin model corresponding to device types 1, 2 and 3 … … M is generated, and a Meta L earning algorithm or strategy is adopted to train the general digital twin model, wherein the general digital twin model not only learns the operation characteristics of the corresponding device, but also learns how quickly to learn device operation data of the energy device type.
After the general digital twin model corresponding to each device is deployed, the sensor network introduces the operation data of the corresponding device into each general digital twin model, and fine-tunes the general digital twin model to obtain the digital twin model corresponding to device 1 … … N. The required data amount is small in the stage, the learning speed is high, the finely adjusted digital twin model is acted on real-time equipment operation data of each corresponding equipment, the digital twin model is used, state information corresponding to the equipment is obtained according to the data of the equipment, and maintenance strategy information corresponding to the equipment is generated based on the state information.
In one embodiment, as shown in fig. 8, the acquisition of the digital twin model is divided into an online process and an offline process, the offline process is a training process of the universal digital twin model, and the online process is a fine tuning process of the universal digital twin model. In the off-line process, a device operation data set corresponding to energy devices belonging to the same energy device type is acquired through a sensor network, the sensor network comprises sensors arranged in the energy devices, and the sensors comprise current sensors, voltage sensors, power sensors, electric/thermal/water meter information, pressure sensors, temperature and humidity sensors, illumination sensors and the like. The trained general digital twin model comprises a power generation model, an energy storage model, a current transformer model, an electrical appliance model and the like.
In the online process, according to the type of the energy equipment to which the energy equipment belongs, a general digital twin model corresponding to the energy equipment is set, second equipment operation data corresponding to the energy equipment is obtained, and the general digital twin model is adjusted based on the second equipment operation data to generate the digital twin model. Fine adjustment of the general digital twin model is carried out by adopting online data so as to adapt to respective characteristics of each actual energy device and generate the digital twin model; and putting the digital twin model into operation, using the digital twin model and acquiring the operation state information corresponding to the energy equipment according to the first equipment operation data to generate maintenance strategy information.
In one embodiment, all trained universal digital twin models are stored in the main node, the type of the energy equipment sent by each workstation is obtained, and the main node is controlled to send the universal digital twin models to the corresponding workstations based on the type of the energy equipment. The master node and the workstations can be controlled by sending control instructions or configuring corresponding operation configuration files on the master node and the workstations.
And when the workstation deploys the universal digital twin model, acquiring the operating data of the second device to adjust the universal digital twin model to generate the digital twin model. And the control workstation sends the running state information output by the digital twin model to the main node so as to generate maintenance strategy information based on the running state information through the main node.
As shown in fig. 9, the master node may act as a decision center and the workstations of the slave nodes may run computers for the digital twin model. An inquiry and storage bidirectional communication channel is established between the workstation and the main node, the storage channel is used for storing, sending and modifying the communication digital twin model, and the inquiry channel is used for transmitting the output result of the model or the input data of the model.
The master node computer has a universal digital twinning model required by each workstation, each workstation can determine the device type through scanning the connected energy devices, the device type is transmitted to the master node through a communication network, and the master node transmits the universal digital twinning model and model parameters to the corresponding workstation.
And when all the workstations are deployed and the universal digital twin model is adjusted, the workstations start to operate the corresponding digital twin models. The information acquired by the sensors is input into the digital twin model by each workstation, corresponding results are obtained, the results are transmitted to the main node through the communication network, and after the main node collects the information from each workstation, the state of the whole system can be evaluated in a sub-device mode or a sub-system mode.
In one embodiment, as shown in fig. 10, the present disclosure provides an energy system maintenance apparatus 10, which includes a model building module 101, a model operating module 102, and a maintenance policy making module 103. The model construction module 101 constructs a digital twin model corresponding to an energy device in an energy system. The model operation module 102 acquires operation state information corresponding to the energy device using the digital twin model and according to first device operation data corresponding to the energy device. The maintenance policy making module 103 generates maintenance policy information corresponding to the energy device based on the operation state information.
In one embodiment, as shown in FIG. 11, model building module 101 includes: a general model generation unit 1011, an apparatus model generation unit 1012, and a model deployment unit 1013. The general model generation unit 1011 generates a general digital twin model corresponding to the energy device type, and acquires a device operation data set corresponding to the energy devices belonging to the same energy device type as a training sample. The general model generation unit 1011 trains the general digital twin model based on the training samples.
The device model generation unit 1012 constructs a digital twin model using the trained general digital twin model. The device model generation unit 1012 sets a general digital twin model corresponding to the energy device according to the type of the energy device to which the energy device belongs. Device model generation section 1012 acquires second device operation data corresponding to the energy device, and generates a digital twin model by adjusting the general digital twin model based on the second device operation data.
The model deployment unit 1013 stores all the trained universal digital twin models in the master node, obtains the types of energy devices sent by each workstation, and controls the master node to send the universal digital twin models to the corresponding workstations based on the types of the energy devices. When the workstation deploys the universal digital twin model, the device model generation unit 1012 acquires second device operation data to adjust the universal digital twin model, and generates a digital twin model.
In one embodiment, the maintenance policy making module 103 controls the workstation to send the operation state information output by the digital twin model to the master node, so as to generate the maintenance policy information based on the operation state information by the master node.
The model operation module 102 obtains first device operation data collected by a sensor network corresponding to the energy device. The device model generation unit 1012 acquires the second device operation data collected by the sensor network. The sensor network includes sensors for collecting first device operational data and second device operational data.
The general model generation unit 1011 trains a general digital twin model based on training samples using a preset machine learning algorithm, which includes a Meta learning Meta L earning algorithm and the like.
The universal digital twin model comprises at least one sub-model; the general model generation unit 1011 acquires modeling information corresponding to the energy device type, the modeling information including: device attributes, device functions, device characteristic information, and the like. The general model generation unit 1011 generates sub models according to the modeling information, wherein the sub models comprise one or more of an equipment state identification model, a fault model, an equipment service life model, an equipment operation transient model and an equipment operation steady-state model. The digital twin model includes device information corresponding to the energy device, at least one sub model.
In one embodiment, the model operation module 1013 inputs the first device operation data into a corresponding digital twin model, and performs simulation operation through the digital twin model to obtain operation state information, where the operation state information includes at least one of historical, current, and predicted operation state information. The maintenance policy making module 1013 evaluates the energy device based on the operation state information, and generates maintenance policy information according to the evaluation result.
In one embodiment, as shown in fig. 12, the present disclosure provides an energy system maintenance system comprising an energy system 120 and a digital twinning system 121; energy system 120 includes energy device 1201 data twin system includes energy system maintenance 1211 as in any of the embodiments above.
Fig. 13 is a block schematic diagram of another embodiment of an energy system maintenance device according to the present disclosure. As shown in fig. 13, the apparatus may include a memory 131, a processor 132, a communication interface 133, and a bus 134. The memory 131 is used for storing instructions, the processor 132 is coupled to the memory 131, and the processor 132 is configured to implement the energy system maintenance method described above based on the instructions stored in the memory 131.
The memory 131 may be a high-speed RAM memory, a non-volatile memory (non-volatile memory), or the like, and the memory 131 may be a memory array. The storage 131 may also be partitioned, and the blocks may be combined into virtual volumes according to certain rules. Processor 132 may be a central processing unit CPU, or an application specific integrated circuit asic (application specific integrated circuit), or one or more integrated circuits configured to implement the energy system maintenance methods of the present disclosure.
In one embodiment, the present disclosure provides a computer-readable storage medium storing computer instructions that, when executed by a processor, implement an energy system maintenance method as in any of the above embodiments.
The energy system maintenance method, the device, the system and the storage medium provided in the above embodiments construct a digital twin model corresponding to an energy device in an energy system, obtain operation state information corresponding to the energy device using the digital twin model and according to first device operation data corresponding to the energy device, and generate maintenance policy information corresponding to the energy device based on the operation state information; the digital twin model can be used for providing maintenance strategies related to energy equipment for designers, manufacturers and maintenance engineers, and higher-quality products and more effective maintenance results are obtained; system variables which cannot be acquired by a conventional sensor can be acquired, and the reason and the degree of equipment abnormity can be acquired in advance; the product quality of the energy equipment can be improved, and more efficient predictive maintenance of the energy equipment is provided; the workload in the equipment maintenance process is reduced, the working efficiency is improved, and the flexibility and the convenience of the energy system maintenance are improved; the constraint of the conventional sensor in the data acquisition process can be broken through; by training the same type of equipment, model fine tuning is performed on actual data of respective physical models after model deployment, the speed of model acquisition is increased, the time consumed by model training is reduced, a customized model is realized, and the use experience of a user is improved.
The method and system of the present disclosure may be implemented in a number of ways. For example, the methods and systems of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order for the steps of the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above unless specifically stated otherwise. Further, in some embodiments, the present disclosure may also be embodied as programs recorded in a recording medium, the programs including machine-readable instructions for implementing the methods according to the present disclosure. Thus, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.
The description of the present disclosure has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the disclosure in the form disclosed. Many modifications and variations will be apparent to practitioners skilled in this art. The embodiment was chosen and described in order to best explain the principles of the disclosure and the practical application, and to enable others of ordinary skill in the art to understand the disclosure for various embodiments with various modifications as are suited to the particular use contemplated.
Claims (27)
1. An energy system maintenance method, comprising:
constructing a digital twin model corresponding to an energy device in an energy system;
acquiring operating state information corresponding to the energy device using the digital twin model and according to first device operating data corresponding to the energy device;
and generating maintenance strategy information corresponding to the energy equipment based on the operation state information.
2. The method of claim 1, the constructing a digital twin model corresponding to an energy device in an energy system comprising:
generating a universal digital twin model corresponding to the energy device type;
acquiring an equipment operation data set corresponding to energy equipment belonging to the same energy equipment type as a training sample;
training the universal digital twin model based on the training samples;
and constructing the digital twin model by using the trained general digital twin model.
3. The method of claim 2, the constructing the digital twin model using the trained universal digital twin model comprising:
setting the universal digital twin model corresponding to the energy equipment according to the type of the energy equipment to which the energy equipment belongs;
adjusting the universal digital twin model based on second device operational data corresponding to the energy device to generate the digital twin model.
4. The method of claim 3, further comprising:
storing all the trained universal digital twin models in a main node;
the energy equipment type sent by each workstation is obtained, and the main node is controlled to send the universal digital twin model to the corresponding workstation based on the energy equipment type;
and when the workstation deploys the universal digital twin model, controlling the workstation to adjust the universal digital twin model by using the second equipment operation data to generate the digital twin model.
5. The method of claim 4, further comprising:
and controlling the workstation to send the running state information output by the digital twin model to the main node, and controlling the main node to generate the maintenance strategy information based on the running state information.
6. The method of claim 3, further comprising:
and controlling the digital twin model to acquire the first equipment operation data and the second equipment operation data acquired by the sensor network corresponding to the energy equipment.
7. The method of claim 6, wherein,
the sensor network includes: a sensor for acquiring the first device operational data and the second device operational data.
8. The method of claim 2, the training the universal digital twin model based on the training samples comprising:
and training the universal digital twin model by using a preset machine learning algorithm and based on the training samples, wherein the machine learning algorithm comprises a Meta-learning Meta L earning algorithm.
9. The method of claim 2, wherein the generic digital twin model comprises: at least one sub-model; the generating of the universal digital twin model corresponding to the energy device type includes:
obtaining modeling information corresponding to the type of the energy equipment; wherein the modeling information comprises: device attributes, device functions, and device feature information;
generating the sub-model according to the modeling information; wherein the sub-models comprise: one or more of a device state identification model, a fault model, a device lifetime model, a device operational transient model, a device operational steady state model.
10. The method of claim 9, wherein,
the digital twinning model comprises: device information corresponding to an energy device, at least one of the submodels.
11. The method of claim 1, the obtaining operational status information corresponding to the energy device from the device operational data using the digital twin model comprising:
inputting the first equipment operation data into the corresponding digital twin model, and performing simulation operation through the digital twin model to obtain the operation state information;
wherein the operating state information includes at least one of historical, current, and predicted operating state information.
12. The method of claim 1, the generating maintenance policy information corresponding to the energy device based on the operating state information comprising:
and evaluating the energy equipment based on the running state information, and generating the maintenance strategy information according to the evaluation and processing result.
13. An energy system maintenance device, comprising:
the model building module is used for building a digital twin model corresponding to energy equipment in the energy system;
the model operation module is used for acquiring operation state information corresponding to the energy equipment by using the digital twin model according to first equipment operation data corresponding to the energy equipment;
and the maintenance strategy making module is used for generating maintenance strategy information corresponding to the energy equipment based on the operation state information.
14. The apparatus of claim 13, wherein,
the model building module comprises:
a general model generation unit for generating a general digital twin model corresponding to the energy device type; acquiring an equipment operation data set corresponding to energy equipment belonging to the same energy equipment type as a training sample; training the universal digital twin model based on the training samples;
and the equipment model generating unit is used for constructing the digital twin model by utilizing the trained general digital twin model.
15. The apparatus of claim 14, wherein,
the device model generating unit is specifically configured to set the universal digital twin model corresponding to the energy device according to the type of the energy device to which the energy device belongs; adjusting the universal digital twin model based on second device operational data corresponding to the energy device to generate the digital twin model.
16. The apparatus of claim 15, wherein,
the model building module comprises:
the model deployment unit is used for storing all the trained universal digital twin models in the main node; the energy equipment type sent by each workstation is obtained, and the main node is controlled to send the universal digital twin model to the corresponding workstation based on the energy equipment type;
and the equipment model generating unit is used for controlling the workstation to adjust the universal digital twin model by using the second equipment operation data to generate the digital twin model when the workstation deploys the universal digital twin model.
17. The apparatus of claim 16, wherein,
the maintenance strategy making module is further configured to control the workstation to send the operation state information output by the digital twin model to the master node, and control the master node to generate the maintenance strategy information based on the operation state information.
18. The apparatus of claim 15, wherein,
the model operation module is further configured to control the digital twin model to obtain the first device operation data and the second device operation data acquired by the sensor network corresponding to the energy device.
19. The apparatus of claim 18, wherein,
the sensor network includes: a sensor for acquiring the first device operational data and the second device operational data.
20. The apparatus of claim 14, wherein,
the general model generation unit is further used for training the general digital twin model by using a preset machine learning algorithm and based on the training samples, wherein the machine learning algorithm comprises a Meta learning Meta L earning algorithm.
21. The apparatus of claim 14, wherein the universal digital twin model comprises: at least one sub-model;
the general model generating unit is further used for acquiring modeling information corresponding to the type of the energy equipment; wherein the modeling information comprises: device attributes, device functions, and device feature information; generating the sub-model according to the modeling information; wherein the sub-models comprise: one or more of a device state identification model, a fault model, a device lifetime model, a device operational transient model, a device operational steady state model.
22. The apparatus of claim 21, wherein,
the digital twinning model comprises: device information corresponding to an energy device, at least one of the submodels.
23. The apparatus of claim 13, wherein,
the model operation module is used for inputting the first equipment operation data into the corresponding digital twin model, and performing simulation operation through the digital twin model to obtain the operation state information; wherein the operating state information includes at least one of historical, current, and predicted operating state information.
24. The apparatus of claim 13, wherein,
and the maintenance strategy making module is used for evaluating the energy equipment based on the running state information and generating the maintenance strategy information according to the evaluation processing result.
25. An energy system maintenance system, comprising:
energy systems and digital twinning systems; the data twinning system comprises: an energy system maintenance device according to any one of claims 13 to 24.
26. An energy system maintenance device, comprising:
a memory; and a processor coupled to the memory, the processor configured to perform the method of any of claims 1-12 based on instructions stored in the memory.
27. A computer-readable storage medium having stored thereon computer instructions for execution by a processor to perform the method of any one of claims 1 to 12.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010454661.7A CN111476386A (en) | 2020-05-26 | 2020-05-26 | Energy system maintenance method, device, system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010454661.7A CN111476386A (en) | 2020-05-26 | 2020-05-26 | Energy system maintenance method, device, system and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111476386A true CN111476386A (en) | 2020-07-31 |
Family
ID=71760156
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010454661.7A Pending CN111476386A (en) | 2020-05-26 | 2020-05-26 | Energy system maintenance method, device, system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111476386A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112993338A (en) * | 2021-03-04 | 2021-06-18 | 上海电气集团股份有限公司 | Monitoring method, device and equipment of energy storage system |
| CN113010950A (en) * | 2021-03-15 | 2021-06-22 | 珠海格力智能装备有限公司 | Factory layout processing method, device, storage medium and processor |
| CN114023132A (en) * | 2021-11-03 | 2022-02-08 | 中国人民解放军火箭军工程大学 | Maintenance training system and method based on digital twins |
| WO2022209047A1 (en) * | 2021-03-31 | 2022-10-06 | オムロン株式会社 | Information management method and information management system |
-
2020
- 2020-05-26 CN CN202010454661.7A patent/CN111476386A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112993338A (en) * | 2021-03-04 | 2021-06-18 | 上海电气集团股份有限公司 | Monitoring method, device and equipment of energy storage system |
| CN113010950A (en) * | 2021-03-15 | 2021-06-22 | 珠海格力智能装备有限公司 | Factory layout processing method, device, storage medium and processor |
| WO2022209047A1 (en) * | 2021-03-31 | 2022-10-06 | オムロン株式会社 | Information management method and information management system |
| CN114023132A (en) * | 2021-11-03 | 2022-02-08 | 中国人民解放军火箭军工程大学 | Maintenance training system and method based on digital twins |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111476386A (en) | Energy system maintenance method, device, system and storage medium | |
| JP5511698B2 (en) | Air conditioner linkage control system, air conditioner linkage control method, and air conditioner linkage control program | |
| EP2682914A1 (en) | Energy management method and system thereof, and gui method | |
| US20140032169A1 (en) | Systems and methods for improving control system reliability | |
| ATE450830T1 (en) | SENSOR FAULT DIAGNOSIS AND PREDICTION USING A COMPONENT MODEL AND TIMESCALE ORTHOGONAL DEVELOPMENTS | |
| US20140031957A1 (en) | Systems and methods for control reliability operations | |
| US20140032470A1 (en) | Systems and methods for control reliability operations | |
| US20140032172A1 (en) | Systems and methods for health assessment of a human-machine interface (hmi) device | |
| Guo et al. | Real-time prediction of remaining useful life and preventive maintenance strategy based on digital twin | |
| KR20210074587A (en) | Communication system apparatus | |
| CN110989403A (en) | Comprehensive energy regulation and control system and control method thereof and server | |
| KR20120080406A (en) | Apparatus and method and for deducting manage plan of building | |
| US20080147361A1 (en) | Methods and apparatus to monitor system health | |
| JP6548862B2 (en) | Method and apparatus for managing power plant operation | |
| Ploennigs et al. | From semantic models to cognitive buildings | |
| Devi et al. | An innovative modular device and wireless control system enabling thermal and pressure sensors using FPGA on real-time fault diagnostics of steam turbine functional deterioration | |
| Yuliza et al. | Modification of Control Oil Feeding with PLC Using Simulation Visual Basic and Neural Network Analysis | |
| Tahmasebi et al. | Smart Integrated Optimization Technique for Large Chilled Water Systems. | |
| US20220377004A1 (en) | Network simulator, network simulation method, and computer-readable recording medium | |
| EP3985577A1 (en) | System and method for optimal drive configuration using machine learning | |
| EP4327228A1 (en) | Methods and systems for operating industrial control loops | |
| JP2023002962A (en) | Information processing apparatus, and program and method for generating model | |
| CN114003990A (en) | Water digital integrated twin control system | |
| Azirah et al. | A data-driven prognostic model for industrial equipment using time series prediction methods | |
| CN117314370A (en) | Intelligent energy-based data cockpit system and implementation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |