CN118607150B - Edge optimization method and system for digital twin space modeling of power distribution network - Google Patents
Edge optimization method and system for digital twin space modeling of power distribution network Download PDFInfo
- Publication number
- CN118607150B CN118607150B CN202411075232.3A CN202411075232A CN118607150B CN 118607150 B CN118607150 B CN 118607150B CN 202411075232 A CN202411075232 A CN 202411075232A CN 118607150 B CN118607150 B CN 118607150B
- Authority
- CN
- China
- Prior art keywords
- distillation
- power distribution
- equipment
- digital twin
- distribution network
- 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.)
- Active
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 119
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000004821 distillation Methods 0.000 claims abstract description 267
- 238000012545 processing Methods 0.000 claims abstract description 69
- 230000004044 response Effects 0.000 claims abstract description 22
- 238000013140 knowledge distillation Methods 0.000 claims abstract description 16
- 230000007613 environmental effect Effects 0.000 claims description 83
- 238000010219 correlation analysis Methods 0.000 claims description 18
- 230000000007 visual effect Effects 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000013138 pruning Methods 0.000 claims description 10
- 230000009467 reduction Effects 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 6
- 238000009877 rendering Methods 0.000 claims description 6
- 230000005540 biological transmission Effects 0.000 abstract description 7
- 230000000694 effects Effects 0.000 abstract description 7
- 230000006870 function Effects 0.000 description 16
- 230000008569 process Effects 0.000 description 6
- 230000000875 corresponding effect Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 238000012512 characterization method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000012876 topography Methods 0.000 description 2
- 238000003646 Spearman's rank correlation coefficient Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
- 238000002759 z-score normalization Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an edge optimization method and system for digital twin space modeling of a power distribution network, which relate to the field of data processing, and the method comprises the following steps: the method comprises the steps of combining environment characteristic information, equipment characteristic information and running state information of power distribution equipment to construct a digital twin teacher model of a power distribution network, obtaining a knowledge distillation edge optimization model, carrying out distillation processing on the environment characteristic information according to a first distillation optimization channel, obtaining an environment distillation result, carrying out distillation processing on the equipment characteristic information by a second distillation optimization channel, obtaining an equipment distillation result, carrying out distillation processing on the running state information by a third distillation optimization channel, obtaining a running distillation result, and obtaining a digital twin student model. The method solves the technical problems of low operation efficiency of the power distribution network caused by delay of data transmission and slow response speed of the edge of the power distribution network, and achieves the technical effect of improving the operation efficiency of the power distribution network by constructing digital twin space modeling to carry out edge optimization on the power distribution network.
Description
Technical Field
The application relates to the field of data processing, in particular to an edge optimization method and an edge optimization system for digital twin space modeling of a power distribution network.
Background
With the development of modern society, the power industry is being reformed and changed, and particularly, the field of power distribution networks, which refers to a power transmission and distribution system from a power source to end users, and is responsible for transmitting power from a power plant to the end users such as a user family, an enterprise, and the like. Along with the growth of power demand and the development of a power system, the edge planning and optimization of a power distribution network become particularly important, and the technical problems of low operation efficiency of the power distribution network caused by delay of edge data transmission and slow response speed exist in the power distribution network nowadays.
Disclosure of Invention
According to the edge optimization method and system for the digital twin space modeling of the power distribution network, the technical problems of low operation efficiency of the power distribution network caused by delay and slow response speed of edge data transmission of the power distribution network are solved, the edge optimization of the power distribution network is realized by constructing the digital twin space modeling, and the technical effect of improving the operation efficiency of the power distribution network is achieved.
The application provides an edge optimization method for digital twin space modeling of a power distribution network, which is applied to an edge optimization system for digital twin space modeling of the power distribution network, and comprises the following steps: acquiring environmental characteristic information of a power distribution network and equipment characteristic information of power distribution equipment in the power distribution network respectively;
combining the environment characteristic information, the equipment characteristic information and the dynamically monitored running state information of the power distribution equipment to construct a digital twin teacher model of the power distribution network;
Obtaining a knowledge distillation edge optimization model, wherein the knowledge distillation edge optimization model is used for carrying out distillation optimization on the digital twin teacher model, and comprises a first distillation optimization channel, a second distillation optimization channel and a third distillation optimization channel;
The first distillation optimization channel carries out distillation treatment on the environmental characteristic information based on a predetermined relation distillation logic to obtain an environmental distillation result;
The second distillation optimization channel carries out distillation treatment on the equipment characteristic information based on a preset characteristic distillation logic to obtain an equipment distillation result;
the third distillation optimization channel performs distillation treatment on the operation state information based on a preset response distillation logic to obtain an operation distillation result;
And obtaining a digital twin student model according to the environmental distillation result, the equipment distillation result and the operation distillation result, wherein the digital twin student model characterizes a distillation optimization model of the digital twin teacher model.
In a possible implementation manner, a digital twin teacher model of the power distribution network is constructed by combining the environment characteristic information, the equipment characteristic information and the dynamically monitored running state information of the power distribution equipment, and the following processing is executed: constructing an initial digital twin teacher model of the power distribution network based on the equipment characteristic information and the running state information;
extracting a first historical record in a historical power distribution network record;
Performing correlation analysis on the first historical environmental characteristic information in the first historical record and a first historical equipment state index to obtain a first historical correlation analysis result;
and carrying out visual representation on the first historical correlation analysis result by using a Manhattan distance, and rendering the visual representation result to the initial digital twin teacher model to obtain the digital twin teacher model.
In a possible implementation manner, the first distillation optimization channel performs distillation processing on the environmental characteristic information based on a predetermined relation distillation logic to obtain an environmental distillation result, and performs the following processing: extracting a predetermined relationship similarity probability distribution in the predetermined relationship distillation logic, wherein the predetermined relationship similarity probability distribution is used for performing distillation processing on the environmental characteristic information;
wherein the expression of the predetermined relationship similarity probability distribution is as follows: ; wherein, For characterizing the similarity divergence of the relationship between the environmental characteristic information and the environmental distillation result toThe environment characteristic information is subjected to dimension reduction processing by taking 0 as a constraint,Characterizing the divergence of the similarity of the relationship between the environmental characteristic information and the environmental distillation results,Characterizing environmental feature points in the environmental feature informationState index with the power distribution equipmentIs a function of the probability of similarity of the relationships,Characterizing environmental feature points in the environmental distillation resultsState index with the power distribution equipmentIs a relationship similarity probability of (1). In a possible implementation, the following process is performed: the running state information comprises a voltage parameter, a current parameter and a power parameter;
Weighting calculation is carried out on the normalization processing results of the voltage parameter, the current parameter and the power parameter, and a real-time state index of the power distribution equipment is obtained;
the real-time state index is used for evaluating faults of the power distribution network and carrying out early warning.
In a possible implementation manner, the second distillation optimization channel performs distillation processing on the equipment characteristic information based on predetermined characteristic distillation logic to obtain equipment distillation results, and performs the following processing: the device characteristic information comprises a plurality of device physical characteristic parameters and a plurality of device electrical characteristic parameters;
Performing weight coefficient distribution on the physical characteristic parameters of the plurality of equipment and the electrical characteristic parameters of the plurality of equipment to obtain equipment characteristic weight distribution results;
The physical characteristic parameters and the electrical characteristic parameters of the plurality of devices are arranged in a descending order according to the device characteristic weight distribution result, and a device characteristic descending order list is obtained;
And pruning the features of the preset last bit proportion threshold value in the equipment feature descending list based on the preset feature distillation logic to obtain the equipment distillation result.
In a possible implementation manner, the third distillation optimization channel performs distillation processing on the operation state information based on a predetermined response distillation logic to obtain an operation distillation result, and performs the following processing: extracting a predetermined student accuracy probability in the predetermined response distillation logic;
reading a predetermined distillation loss function and combining the predetermined student accuracy probability and the real-time state index to obtain the operation distillation result, wherein the expression of the predetermined distillation loss function is as follows: ; wherein, The predetermined distillation loss function is characterized in that,Characterizing the real-time state index obtained by the digital twin teacher modelStudent real-time state index obtained by the digital twin student modelThe degree of divergence between the two,The temperature coefficient is characterized by the fact that,Characterizing the digital twin teacher modelThe first of the real-time state indexesA real-time state index. In a possible implementation, the following process is performed: acquiring an index difference between the student real-time state index and the real-time state index;
judging whether the index difference reaches a preset difference range;
and if the index difference does not reach the preset difference range, reducing and adjusting the preset last bit proportion threshold value.
The application also provides an edge optimization system for modeling the digital twin space of the power distribution network, which comprises the following steps: the information acquisition module is used for acquiring environmental characteristic information of the power distribution network and equipment characteristic information of power distribution equipment in the power distribution network respectively;
the first model building module is used for building a digital twin teacher model of the power distribution network by combining the environment characteristic information, the equipment characteristic information and the dynamically monitored running state information of the power distribution equipment;
the distillation optimization module is used for acquiring a knowledge distillation edge optimization model which is used for carrying out distillation optimization on the digital twin teacher model, wherein the knowledge distillation edge optimization model comprises a first distillation optimization channel, a second distillation optimization channel and a third distillation optimization channel;
The first distillation processing module is used for the first distillation optimization channel to carry out distillation processing on the environmental characteristic information based on a predetermined relation distillation logic to obtain an environmental distillation result;
The second distillation processing module is used for performing distillation processing on the equipment characteristic information based on a preset characteristic distillation logic by the second distillation optimization channel to obtain equipment distillation results;
The third distillation processing module is used for performing distillation processing on the running state information by the third distillation optimization channel based on a preset response distillation logic to obtain a running distillation result;
The second model construction module is used for obtaining a digital twin student model according to the environment distillation result, the equipment distillation result and the operation distillation result, and the digital twin student model characterizes a distillation optimization model of the digital twin teacher model.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
The application provides an edge optimization method and an edge optimization system for digital twin space modeling of a power distribution network, relates to the technical field of data processing, solves the technical problems of low operation efficiency of the power distribution network caused by slow edge data transmission delay and response speed of the power distribution network, and achieves the technical effect of improving the operation efficiency of the power distribution network by constructing digital twin space modeling to perform edge optimization on the power distribution network.
Drawings
In order to more clearly illustrate the technical solution of the embodiments of the present application, the following description will briefly refer to the accompanying drawings of the embodiments of the present application, in which flowcharts are used to illustrate operations performed by systems according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in order precisely. Rather, the various steps may be processed in reverse order or simultaneously, as desired. Also, other operations may be added to or removed from these processes.
Fig. 1 is a schematic flow chart of an edge optimization method for digital twin space modeling of a power distribution network according to an embodiment of the present application;
Fig. 2 is a schematic structural diagram of an edge optimization system for digital twin space modeling of a power distribution network according to an embodiment of the present application.
Detailed Description
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be described in further detail with reference to the accompanying drawings, and the described embodiments should not be construed as limiting the present application, and all other embodiments obtained by those skilled in the art without making any inventive effort are within the scope of the present application.
In the following description, reference is made to "some embodiments" which describe a subset of all possible embodiments, but it is to be understood that "some embodiments" may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict, the term "first\second" being referred to merely as distinguishing between similar objects and not representing a particular ordering for the objects. The terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or server that comprises a list of steps or elements is not necessarily limited to those steps or elements that are expressly listed or inherent to such process, method, article, or apparatus, but may include other steps or modules that may not be expressly listed or inherent to such process, method, article, or apparatus, and unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. The terminology used herein is for the purpose of describing embodiments of the application only.
The embodiment of the application provides an edge optimization method for digital twin space modeling of a power distribution network, which is applied to an edge optimization system for digital twin space modeling of the power distribution network, as shown in fig. 1, and comprises the following steps:
step A100, respectively acquiring environmental characteristic information of a power distribution network and equipment characteristic information of power distribution equipment in the power distribution network;
Environmental characteristics such as climate, topography, geology, electromagnetic environment in the distribution network are acquired through various sensors and monitoring equipment, environmental characteristic information of the distribution network is obtained, further, equipment characteristics such as running states, performance parameters, fault records and the like of the distribution equipment in the distribution network are acquired through the sensors and the monitoring system installed on the equipment, equipment characteristic information of the distribution equipment is obtained, and then the environmental characteristic information of the distribution network and the equipment characteristic information of the distribution equipment can be further processed and analyzed, so that useful characteristics are extracted, a digital twin model is constructed, the running of the distribution network is optimized, and the power supply reliability and economical efficiency are improved.
Executing step A200, and constructing a digital twin teacher model of the power distribution network by combining the environment characteristic information, the equipment characteristic information and the dynamically monitored running state information of the power distribution equipment;
In a possible implementation manner, the step a200 further includes a step a210 of constructing an initial digital twin teacher model of the power distribution network based on the device feature information and the operation state information;
In one possible implementation, step a210 further includes step a211, where the operating state information includes a voltage parameter, a current parameter, and a power parameter; step A212 is executed, and the normalization processing results of the voltage parameter, the current parameter and the power parameter are weighted and calculated to obtain a real-time state index of the power distribution equipment;
the real-time state index is used for evaluating faults of the power distribution network and carrying out early warning.
The obtained environmental characteristic information and the equipment characteristic information are combined with the dynamically monitored running state information to be used as basic data for constructing a digital twin classroom model of the power distribution network, the running state information is data for evaluating the health state of the power distribution network and predicting potential faults, the voltage parameter, the current parameter and the power parameter of the power distribution equipment in the power distribution network are collected and then normalized, and the voltage parameter, the current parameter and the power parameter are required to be converted to the same scale so as to carry out subsequent comparison and calculation, the normalization processing can be carried out through normalization processing methods such as Z-score normalization, the importance degree of the equipment state is calculated according to the voltage parameter, the current parameter and the power parameter after the normalization processing, the weighting calculation can integrate a plurality of parameters, namely the voltage parameter, the current parameter and the power parameter into a single real-time state index, the real-time state index can be used for reflecting the integral running state of the power distribution equipment at a certain moment, and can be a dynamically changed numerical value, wherein the real-time state index is used for evaluating the power distribution equipment and warning the fault state when the power distribution equipment is low in real-time, and the real-time state is monitored, and the fault state is timely monitored.
Further, the equipment characteristic information and the running state information are integrated to form a complete data set, a digital twin model of the power distribution network is built by utilizing a proper modeling method and a proper modeling tool based on the integrated data, and verification of an initial digital twin model is carried out by comparing the digital twin model with actual running data, carrying out simulation experiments and the like, so that the running condition of the power distribution network can be accurately simulated, and the structure, equipment and running state of the power distribution network can be accurately reflected by the initial digital twin model.
Executing step A220, and extracting a first historical record in the historical power distribution network records; executing step A230, performing correlation analysis on the first historical environmental characteristic information in the first historical record and the first historical equipment state index to obtain a first historical correlation analysis result;
The method comprises the steps of firstly extracting a first historical record from a historical power distribution network record database, wherein the first historical record can comprise a plurality of pieces of historical environment characteristic information and equipment historical state index data so as to carry out subsequent analysis, the historical environment characteristic information can comprise weather data such as temperature, humidity, wind speed and air pressure, geographical data such as topography and geology, the equipment historical state index is calculated based on real-time operation state information and reflects the overall operation state of equipment at a certain historical moment, then, the method such as a pearson correlation coefficient, a spearman rank correlation coefficient and the like can be used for analyzing and quantifying the correlation between the environment characteristic information and the equipment state index, a first historical correlation analysis result corresponding to the first historical record can be obtained through correlation analysis, the first historical correlation analysis result comprises the correlation between the historical environment characteristic information and the equipment historical state index, and the nature and strength of the correlation, and the weather factors (such as high temperature and high humidity) can be positively correlated with the reduction of the power distribution equipment state index, and the normal operation of the equipment can be influenced by the environment factors.
And executing a step A240, performing visual representation on the first historical correlation analysis result with a Manhattan distance, and rendering the visual representation result to the initial digital twin teacher model to obtain the digital twin teacher model.
The method comprises the steps of regarding environmental characteristic information and equipment state indexes in a first historical correlation analysis result as points in a multidimensional space based on Manhattan distance, measuring the distance between the points by using the Manhattan distance, further visually processing the Manhattan distance calculation result by using a scatter diagram, a thermodynamic diagram and other visualization tools, and displaying the first historical correlation analysis result in a graphical mode, wherein the position or color of each point can represent the correlation strength and direction between the environmental characteristic information and the equipment state indexes, so that a visual characterization result is obtained, and rendering the visual characterization result into an initial digital twin teacher model obtained by constructing by adding a corresponding visual layer or component in the initial digital twin teacher model, so that the initial digital twin teacher model can intuitively display the historical correlation analysis result, and further achieve the purpose of combining the data analysis result with the physical structure of the initial digital twin teacher to form a complete and accurate digital twin model to represent the digital twin teacher, and further provide more complete and accurate operation condition of a power distribution network, and more comprehensive operation and prediction and operation condition of a power distribution network.
Executing step A300, obtaining a knowledge distillation edge optimization model, wherein the knowledge distillation edge optimization model is used for carrying out distillation optimization on the digital twin teacher model, and comprises a first distillation optimization channel, a second distillation optimization channel and a third distillation optimization channel;
In order to optimize the performance of the digital twin teacher model in an edge computing environment, a knowledge distillation edge optimization model can be constructed, wherein the knowledge distillation edge optimization model comprises three main distillation optimization channels: the first distillation optimizing channel, the second distillation optimizing channel and the third distillation optimizing channel, and each channel respectively carries out distillation treatment on environment characteristic information, equipment characteristic information and running state information based on different preset distillation logics.
Executing a step A400, wherein the first distillation optimization channel carries out distillation treatment on the environmental characteristic information based on a predetermined relation distillation logic to obtain an environmental distillation result;
in a possible implementation manner, the step a400 further includes a step a410 of extracting a predetermined relationship similarity probability distribution in the predetermined relationship distillation logic, where the predetermined relationship similarity probability distribution is used for performing distillation processing on the environmental feature information;
before extracting a predetermined relationship similarity probability distribution in a predetermined relationship distillation logic, the predetermined relationship distillation logic is firstly required to be determined, the predetermined relationship distillation logic is based on correlation between environment characteristic information and equipment states, interaction between different environment characteristics and the like, and the predetermined relationship similarity probability distribution in the predetermined relationship distillation logic is based on important environment characteristic information in a digital twin teacher model and relative relationship strength between the environment characteristic information and the interaction between different environment characteristics, wherein the expression of the predetermined relationship similarity probability distribution is as follows: ; wherein, For characterizing the similarity divergence of the relationship between the environmental characteristic information and the environmental distillation result toThe environment characteristic information is subjected to dimension reduction processing by taking 0 as a constraint,Characterizing the divergence of the similarity of the relationship between the environmental characteristic information and the environmental distillation results,Characterizing environmental feature points in the environmental feature informationState index with the power distribution equipmentIs a function of the probability of similarity of the relationships,Characterizing environmental feature points in the environmental distillation resultsState index with the power distribution equipmentIs a relationship similarity probability of (1). The predetermined relationship similarity probability distribution in the predetermined relationship distillation logic is used for the similarity probability distribution between different environmental characteristic information or between the environmental characteristic information and the equipment state, and the predetermined relationship similarity probability distribution is applied to the distillation processing of the environmental characteristic information, so that the probability distribution can be used for guiding the digital twin student model to focus on the information similar to the key environmental characteristic in the teacher model in the distillation process, the student model can keep the characteristic representation and relationship structure similar to the teacher model when receiving and learning the environmental characteristic information, and the performance and applicability of the student model in the edge computing environment can be improved.
Step A500 is executed, and the second distillation optimization channel performs distillation processing on the equipment characteristic information based on a preset characteristic distillation logic to obtain equipment distillation results; the predetermined characteristic distillation logic focuses on extracting key characteristics such as types, specifications, performance parameters and the like of equipment, and distillation processing is performed on equipment characteristic information based on the predetermined characteristic distillation logic through a second distillation optimization channel.
In a possible implementation, the step a500 further includes a step a510, where the device characteristic information includes a plurality of device physical characteristic parameters and a plurality of device electrical characteristic parameters; step A520 is executed, wherein weight coefficient distribution is performed on the physical characteristic parameters of the plurality of devices and the electrical characteristic parameters of the plurality of devices, so as to obtain a device characteristic weight distribution result; executing step a530, namely performing descending order arrangement on the physical characteristic parameters of the plurality of devices and the electrical characteristic parameters of the plurality of devices according to the device characteristic weight distribution result to obtain a device characteristic descending order list;
Because the effective processing of the equipment characteristic information is critical to the improvement of the performance of the model, and a plurality of physical characteristic parameters such as size, material, weight and the like and a plurality of electrical characteristic parameters such as voltage, current, power and the like can be included in the equipment characteristic information, further the distribution of weight coefficients based on the experience of field experts and the statistical analysis of historical data is carried out on the physical characteristic parameters and the electrical characteristic parameters of the equipment, a corresponding weight value is distributed to each parameter according to the importance of the parameter in the prediction of the running state of the equipment, the weight value reflects the relative importance of the parameter in the model, and then the physical characteristic parameters and the electrical characteristic parameters of the equipment are arranged in descending order according to the equipment characteristic weight distribution result, namely the most important parameters are arranged at the forefront of a list, so that the distribution is carried out, the follow-up model optimization and the distillation process are better called, and the equipment characteristic descending order list is obtained on the basis.
And executing a step A540, and pruning the features with the preset last bit proportion threshold value in the equipment feature descending list based on the preset feature distillation logic to obtain the equipment distillation result.
Firstly, determining the features to be reserved and the features to be pruned according to a preset feature distillation logic, wherein the preset feature distillation logic can be set analytically based on domain knowledge, expert experience or historical data and is used for defining the importance and reserved standard of the features, then in order to reduce the model size, reduce the model calculation complexity, improve the model generalization capability and reduce the risk of overfitting, the features of a preset last bit proportion threshold are set, and an exemplary threshold can be set to 90%, the features with the maximum reserved weight are regarded as the first 90%, the remaining 10% of the features are pruned, the specific operation of pruning can be to directly remove the last bit features from the model or set the weights of the last bit features to zero, so that the influence of the last bit features in the model is reduced to the minimum, and a simplified device feature set, namely a device distillation result is obtained.
Further, since unstructured pruning is efficient but susceptible to accuracy, in one possible implementation, step a540 further includes step a541 of obtaining an index difference between the student real-time status index and the real-time status index; step A542 is executed to determine whether the exponent difference reaches a predetermined difference range; and step A543, if the exponent difference does not reach the predetermined difference range, performing reduction adjustment on the predetermined last bit proportion threshold.
Firstly, the index difference between the student real-time state index and the real-time state index is obtained, the student real-time state index and the real-time state index respectively represent the performance of the student model and the teacher model at specific moments, the performance difference between the student model and the teacher model can be obtained by comparing the student real-time state index with the real-time state index, whether the index difference reaches a preset difference range or not is judged, the preset difference range is a threshold value set according to actual requirements and is used for judging whether the performance difference between the student model and the teacher model is overlarge or not, if the index difference does not reach the preset difference range, the performance of the student model is still acceptable, but still has an optimized space, the preset final position proportion threshold value is reduced and adjusted, meanwhile, a proper adjustment range is required according to the size of the index difference, if the index difference is small, the performance of the student model is close to the teacher model, the threshold value can be moderately reduced, if the index difference is large, the performance of the student model is also has a large lifting space, at the moment, the threshold value can be reduced more, the preset final position proportion threshold value is the quantity of features reserved during pruning processing is reduced, so that key features are simplified, and meanwhile, the performance of the student model can be further reduced by the preset final position proportion feature model is more and the key feature is reduced, and the performance feature is more can be filtered by the threshold value.
Executing a step A600, wherein the third distillation optimization channel performs distillation processing on the operation state information based on a predetermined response distillation logic to obtain an operation distillation result;
In one possible implementation, step a600 further includes step a610, extracting a predetermined student accuracy probability in the predetermined response distillation logic; step a620 is performed, wherein a predetermined distillation loss function is read, and the operation distillation result is obtained by combining the predetermined student accuracy probability and the real-time state index,
Firstly, extracting a predetermined student accuracy probability from predetermined response distillation logic, wherein the predetermined student accuracy probability is obtained by setting based on domain knowledge, expert experience or historical data and can be used for measuring the accuracy of a student model on response prediction and reflecting the expectation of the performance of the student model on a specific task, and further reading a predetermined distillation loss function, wherein the expression of the predetermined distillation loss function is as follows: ; wherein, The predetermined distillation loss function is characterized in that,Characterizing the real-time state index obtained by the digital twin teacher modelStudent real-time state index obtained by the digital twin student modelThe degree of divergence between the two,The temperature coefficient is characterized by the fact that,Characterizing the digital twin teacher modelThe first of the real-time state indexesA real-time state index.
The distillation loss function is an important index for measuring the difference between the student model and the teacher model, can be used for measuring the difference between the probability distribution output by the student model of the digital twin model and the probability distribution output by the teacher model of the digital twin model, and the prediction loss of the digital twin student model, further combines the accuracy probability of a preset student with a real-time state index, and inputs the combined result into the preset distillation loss function together, so that the accuracy and performance of the student model in response prediction can be more comprehensively evaluated, the real-time state index is used for reflecting the performance state of the student model in real-time operation and can be used as an important reference for the performance evaluation of the student model, and finally, the operation distillation result is obtained through calculation of the preset distillation loss function.
And executing the step A700, and obtaining a digital twin student model according to the environmental distillation result, the equipment distillation result and the operation distillation result, wherein the digital twin student model characterizes a distillation optimization model of the digital twin teacher model.
The method comprises the steps of obtaining adaptability of a digital twin student model in different environments by analyzing information of model performance under different environmental conditions provided by an environmental distillation result, adjusting the structure and parameters of the digital twin student model accordingly to improve stability and accuracy of the digital twin student model in various environments, obtaining an optimization effect of the digital twin student model on specific equipment by using a analysis device distillation result, ensuring that the digital twin student model can keep higher performance on target equipment, reducing calculation complexity and resource consumption, evaluating performance of the student model in real time by using an operation distillation result, and obtaining the accuracy and the efficiency of the digital twin student model in simulating the digital twin teacher model by comparing the real-time state index of the digital twin student model with the state index of the digital twin teacher model so as to further adjust the parameters and the structure of the student model.
The method and the device solve the technical problems of low operation efficiency of the power distribution network caused by delay of data transmission and slow response speed of the edge of the power distribution network, realize edge optimization of the power distribution network by constructing digital twin space modeling, and achieve the technical effect of improving the operation efficiency of the power distribution network.
In the above, an edge optimization method for digital twin space modeling of a power distribution network according to an embodiment of the present invention is described in detail with reference to fig. 1. Next, an edge optimization system for digital twin space modeling of a power distribution network according to an embodiment of the present invention will be described with reference to fig. 2.
According to the edge optimization system for the digital twin space modeling of the power distribution network, which is disclosed by the embodiment of the invention, the technical problems of low operation efficiency of the power distribution network caused by delay and slow response speed of edge data transmission of the power distribution network are solved, the edge optimization of the power distribution network is realized by constructing the digital twin space modeling, and the technical effect of improving the operation efficiency of the power distribution network is achieved. An edge optimization system for digital twin space modeling of a power distribution network comprising: an information acquisition module 10, a first model construction module 20, a distillation optimization module 30, a first distillation processing module 40, a second distillation processing module 50, a third distillation processing module 60, and a second model construction module 70.
The information acquisition module 10 is used for respectively acquiring environmental characteristic information of a power distribution network and equipment characteristic information of power distribution equipment in the power distribution network;
the first model building module 20 is configured to build a digital twin teacher model of the power distribution network by combining the environmental characteristic information, the equipment characteristic information and the dynamically monitored running state information of the power distribution equipment;
The distillation optimization module 30 is configured to obtain a knowledge distillation edge optimization model, where the knowledge distillation edge optimization model is configured to perform distillation optimization on the digital twin teacher model, and the knowledge distillation edge optimization model includes a first distillation optimization channel, a second distillation optimization channel, and a third distillation optimization channel;
The first distillation processing module 40 is configured to perform distillation processing on the environmental characteristic information by using the first distillation optimization channel based on a predetermined relationship distillation logic, so as to obtain an environmental distillation result;
a second distillation processing module 50, where the second distillation processing module 50 is configured to perform distillation processing on the device feature information by using the second distillation optimization channel based on a predetermined feature distillation logic, so as to obtain a device distillation result;
The third distillation processing module 60, where the third distillation processing module 60 is configured to perform distillation processing on the running state information by using the third distillation optimization channel based on a predetermined response distillation logic, so as to obtain a running distillation result;
And the second model construction module 70 is used for obtaining a digital twin student model according to the environmental distillation result, the equipment distillation result and the operation distillation result, wherein the digital twin student model characterizes a distillation optimization model of the digital twin teacher model.
Next, the specific configuration of the first model building module 20 will be described in detail. As described above, the digital twin teacher model of the power distribution network is constructed by combining the environmental characteristic information, the device characteristic information and the dynamically monitored operation state information of the power distribution device, and the first model construction module 20 may further include: the third model building unit is used for building an initial digital twin teacher model of the power distribution network based on the equipment characteristic information and the running state information; the first extraction unit is used for extracting a first history record in the history power distribution network records; the first analysis unit is used for carrying out correlation analysis on the first historical environmental characteristic information in the first historical record and the first historical equipment state index to obtain a first historical correlation analysis result; the first rendering unit is used for carrying out visual representation on the first historical correlation analysis result by using a Manhattan distance, and rendering the visual representation result to the initial digital twin teacher model to obtain the digital twin teacher model.
Next, the specific configuration of the first distillation treatment module 40 will be described in detail. As described above, the first distillation optimization channel performs distillation processing on the environmental characteristic information based on a predetermined relationship distillation logic to obtain an environmental distillation result, and the first distillation processing module 40 may further include: the second extraction unit is used for extracting a predetermined relationship similarity probability distribution in the predetermined relationship distillation logic, and the predetermined relationship similarity probability distribution is used for performing distillation processing on the environment characteristic information;
wherein the expression of the predetermined relationship similarity probability distribution is as follows: ; wherein, For characterizing the similarity divergence of the relationship between the environmental characteristic information and the environmental distillation result toThe environment characteristic information is subjected to dimension reduction processing by taking 0 as a constraint,Characterizing the divergence of the similarity of the relationship between the environmental characteristic information and the environmental distillation results,Characterizing environmental feature points in the environmental feature informationState index with the power distribution equipmentIs a function of the probability of similarity of the relationships,Characterizing environmental feature points in the environmental distillation resultsState index with the power distribution equipmentIs a relationship similarity probability of (1).
Next, a specific configuration of the third model building unit will be described in detail. As described above, the third model building unit may further include: the first parameter acquisition unit is used for acquiring the running state information, wherein the running state information comprises a voltage parameter, a current parameter and a power parameter; the first calculation unit is used for carrying out weighted calculation on the normalization processing results of the voltage parameter, the current parameter and the power parameter to obtain a real-time state index of the power distribution equipment;
the real-time state index is used for evaluating faults of the power distribution network and carrying out early warning.
Next, the specific configuration of the second distillation treatment module 50 will be described in detail. As described above, the second distillation optimization channel performs distillation processing on the device characteristic information based on predetermined characteristic distillation logic to obtain a device distillation result, and the second distillation processing module 50 may further include: the second parameter acquisition unit is used for the equipment characteristic information to comprise a plurality of equipment physical characteristic parameters and a plurality of equipment electrical characteristic parameters; the coefficient distribution unit is used for distributing weight coefficients of the plurality of equipment physical characteristic parameters and the plurality of equipment electrical characteristic parameters to obtain equipment characteristic weight distribution results; the descending order arrangement unit is used for descending order of the physical characteristic parameters of the plurality of equipment and the electrical characteristic parameters of the plurality of equipment according to the equipment characteristic weight distribution result to obtain an equipment characteristic descending order list; and the pruning processing unit is used for pruning the features of the preset last bit proportion threshold value in the equipment feature descending list based on the preset feature distillation logic to obtain the equipment distillation result.
Next, the specific configuration of the third distillation treatment module 60 will be described in detail. As described above, the third distillation optimization channel performs distillation processing on the operation state information based on a predetermined response distillation logic to obtain an operation distillation result, and the third distillation processing module 60 may further include: the third extraction unit is used for extracting the accuracy probability of the predetermined student in the predetermined response distillation logic; the first reading unit is used for reading a preset distillation loss function and obtaining the operation distillation result by combining the preset student accuracy probability and the real-time state index, wherein the expression of the preset distillation loss function is as follows: ; wherein, The predetermined distillation loss function is characterized in that,Characterizing the real-time state index obtained by the digital twin teacher modelStudent real-time state index obtained by the digital twin student modelThe degree of divergence between the two,The temperature coefficient is characterized by the fact that,Characterizing the digital twin teacher modelThe first of the real-time state indexesA real-time state index.
Next, the specific configuration of the pruning processing unit will be described in detail. As described above, the pruning processing unit may further include: the second calculation unit is used for obtaining the index difference between the student real-time state index and the real-time state index; the first judging unit is used for judging whether the index difference reaches a preset difference range; and the reduction adjustment unit is used for carrying out reduction adjustment on the preset last bit proportion threshold value if the index difference does not reach the preset difference range.
The edge optimization system for the digital twin space modeling of the power distribution network provided by the embodiment of the invention can execute the edge optimization method for the digital twin space modeling of the power distribution network provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
Although the present application makes various references to certain modules in a system according to embodiments of the present application, any number of different modules may be used and run on a user terminal and/or server, including units and modules that are merely partitioned by functional logic, but are not limited to the above-described partitioning, so long as the corresponding functionality is enabled; in addition, the specific names of the functional units are also only for distinguishing from each other, and are not used to limit the protection scope of the present application.
The above embodiments do not limit the scope of the present application. It will be apparent to those skilled in the art that various modifications, combinations, and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the scope of the present application.
Claims (7)
1. An edge optimization method for digital twin space modeling of a power distribution network, comprising the steps of:
Acquiring environmental characteristic information of a power distribution network and equipment characteristic information of power distribution equipment in the power distribution network respectively;
combining the environment characteristic information, the equipment characteristic information and the dynamically monitored running state information of the power distribution equipment to construct a digital twin teacher model of the power distribution network;
Obtaining a knowledge distillation edge optimization model, wherein the knowledge distillation edge optimization model is used for carrying out distillation optimization on the digital twin teacher model, and comprises a first distillation optimization channel, a second distillation optimization channel and a third distillation optimization channel;
The first distillation optimization channel carries out distillation treatment on the environmental characteristic information based on a predetermined relation distillation logic to obtain an environmental distillation result;
The second distillation optimization channel carries out distillation treatment on the equipment characteristic information based on a preset characteristic distillation logic to obtain an equipment distillation result;
the third distillation optimization channel performs distillation treatment on the operation state information based on a preset response distillation logic to obtain an operation distillation result;
obtaining a digital twin student model according to the environmental distillation result, the equipment distillation result and the operation distillation result, wherein the digital twin student model characterizes a distillation optimization model of the digital twin teacher model;
Combining the environmental characteristic information, the equipment characteristic information and the dynamically monitored running state information of the power distribution equipment to construct a digital twin teacher model of the power distribution network, wherein the digital twin teacher model comprises the following components:
constructing an initial digital twin teacher model of the power distribution network based on the equipment characteristic information and the running state information;
extracting a first historical record in a historical power distribution network record;
Performing correlation analysis on the first historical environmental characteristic information in the first historical record and a first historical equipment state index to obtain a first historical correlation analysis result;
and carrying out visual representation on the first historical correlation analysis result by using a Manhattan distance, and rendering the visual representation result to the initial digital twin teacher model to obtain the digital twin teacher model.
2. The edge optimization method for digital twin space modeling of a power distribution network according to claim 1, wherein the first distillation optimization channel performs distillation processing on the environmental characteristic information based on a predetermined relationship distillation logic to obtain an environmental distillation result, and the method comprises the following steps:
Extracting a predetermined relationship similarity probability distribution in the predetermined relationship distillation logic, wherein the predetermined relationship similarity probability distribution is used for performing distillation processing on the environmental characteristic information;
wherein the expression of the predetermined relationship similarity probability distribution is as follows: ; wherein, For characterizing the similarity divergence of the relationship between the environmental characteristic information and the environmental distillation result toThe environment characteristic information is subjected to dimension reduction processing by taking 0 as a constraint,Characterizing the divergence of the similarity of the relationship between the environmental characteristic information and the environmental distillation results,Characterizing environmental feature points in the environmental feature informationState index with the power distribution equipmentIs a function of the probability of similarity of the relationships,Characterizing environmental feature points in the environmental distillation resultsState index with the power distribution equipmentIs a relationship similarity probability of (1).
3. An edge optimization method for digital twin space modeling of a power distribution network according to claim 2, further comprising:
The running state information comprises a voltage parameter, a current parameter and a power parameter;
Weighting calculation is carried out on the normalization processing results of the voltage parameter, the current parameter and the power parameter, and a real-time state index of the power distribution equipment is obtained;
the real-time state index is used for evaluating faults of the power distribution network and carrying out early warning.
4. An edge optimization method for digital twin space modeling of a power distribution network according to claim 3, wherein the second distillation optimization channel performs distillation processing on the equipment characteristic information based on predetermined characteristic distillation logic to obtain an equipment distillation result, and the method comprises the following steps:
the device characteristic information comprises a plurality of device physical characteristic parameters and a plurality of device electrical characteristic parameters;
Performing weight coefficient distribution on the physical characteristic parameters of the plurality of equipment and the electrical characteristic parameters of the plurality of equipment to obtain equipment characteristic weight distribution results;
The physical characteristic parameters and the electrical characteristic parameters of the plurality of devices are arranged in a descending order according to the device characteristic weight distribution result, and a device characteristic descending order list is obtained;
And pruning the features of the preset last bit proportion threshold value in the equipment feature descending list based on the preset feature distillation logic to obtain the equipment distillation result.
5. An edge optimization method for digital twin space modeling of a power distribution network according to claim 3, wherein the third distillation optimization channel performs distillation processing on the operation state information based on a predetermined response distillation logic to obtain an operation distillation result, and the method comprises:
extracting a predetermined student accuracy probability in the predetermined response distillation logic;
reading a predetermined distillation loss function and combining the predetermined student accuracy probability and the real-time state index to obtain the operation distillation result, wherein the expression of the predetermined distillation loss function is as follows: ；
wherein, The predetermined distillation loss function is characterized in that,Characterizing the real-time state index obtained by the digital twin teacher modelStudent real-time state index obtained by the digital twin student modelThe degree of divergence between the two,The temperature coefficient is characterized by the fact that,Characterizing the digital twin teacher modelThe first of the real-time state indexesA real-time state index.
6. An edge optimization method for digital twin space modeling of a power distribution network as defined in claim 4, further comprising:
Acquiring an index difference between a student real-time state index and a real-time state index;
judging whether the index difference reaches a preset difference range;
and if the index difference does not reach the preset difference range, reducing and adjusting the preset last bit proportion threshold value.
7. An edge optimization system for digital twin space modeling of a power distribution network, the system being configured to implement an edge optimization method for digital twin space modeling of a power distribution network as defined in any one of claims 1-6, comprising:
the information acquisition module is used for acquiring environmental characteristic information of the power distribution network and equipment characteristic information of power distribution equipment in the power distribution network respectively;
the first model building module is used for building a digital twin teacher model of the power distribution network by combining the environment characteristic information, the equipment characteristic information and the dynamically monitored running state information of the power distribution equipment;
the distillation optimization module is used for acquiring a knowledge distillation edge optimization model which is used for carrying out distillation optimization on the digital twin teacher model, wherein the knowledge distillation edge optimization model comprises a first distillation optimization channel, a second distillation optimization channel and a third distillation optimization channel;
The first distillation processing module is used for the first distillation optimization channel to carry out distillation processing on the environmental characteristic information based on a predetermined relation distillation logic to obtain an environmental distillation result;
The second distillation processing module is used for performing distillation processing on the equipment characteristic information based on a preset characteristic distillation logic by the second distillation optimization channel to obtain equipment distillation results;
The third distillation processing module is used for performing distillation processing on the running state information by the third distillation optimization channel based on a preset response distillation logic to obtain a running distillation result;
The second model construction module is used for obtaining a digital twin student model according to the environment distillation result, the equipment distillation result and the operation distillation result, and the digital twin student model characterizes a distillation optimization model of the digital twin teacher model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411075232.3A CN118607150B (en) | 2024-08-07 | 2024-08-07 | Edge optimization method and system for digital twin space modeling of power distribution network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411075232.3A CN118607150B (en) | 2024-08-07 | 2024-08-07 | Edge optimization method and system for digital twin space modeling of power distribution network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118607150A CN118607150A (en) | 2024-09-06 |
| CN118607150B true CN118607150B (en) | 2024-11-05 |
Family
ID=92557362
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411075232.3A Active CN118607150B (en) | 2024-08-07 | 2024-08-07 | Edge optimization method and system for digital twin space modeling of power distribution network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118607150B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117710792A (en) * | 2023-12-27 | 2024-03-15 | 上海人工智能创新中心 | Knowledge distillation method, greening region detection method, electronic device, and storage medium |
| CN117812593A (en) * | 2024-01-09 | 2024-04-02 | 重庆邮电大学 | Intrusion detection method under knowledge distillation-fused group learning architecture |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118134966A (en) * | 2024-03-06 | 2024-06-04 | 西安理工大学 | Knowledge distillation method of twin architecture tracking model based on hierarchical feature fusion |
| CN118017509B (en) * | 2024-04-10 | 2024-07-16 | 国网江苏省电力有限公司南通供电分公司 | Large-scale power distribution network parallel optimization method based on digital twin space |
| CN118381197B (en) * | 2024-06-26 | 2024-10-01 | 国网江苏省电力有限公司南通供电分公司 | Power distribution network fault monitoring method and system based on digital twinning |
-
2024
- 2024-08-07 CN CN202411075232.3A patent/CN118607150B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117710792A (en) * | 2023-12-27 | 2024-03-15 | 上海人工智能创新中心 | Knowledge distillation method, greening region detection method, electronic device, and storage medium |
| CN117812593A (en) * | 2024-01-09 | 2024-04-02 | 重庆邮电大学 | Intrusion detection method under knowledge distillation-fused group learning architecture |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118607150A (en) | 2024-09-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111539515B (en) | Complex equipment maintenance decision method based on fault prediction | |
| CN111459700A (en) | Method and apparatus for diagnosing device failure, diagnostic device, and storage medium | |
| CN112116198B (en) | Data-driven process industrial state perception network key node screening method | |
| CN114169254B (en) | Abnormal energy consumption diagnosis method and system based on short-term building energy consumption prediction model | |
| CN113688558A (en) | Automobile driving condition construction method and system based on large database samples | |
| CN115438726A (en) | Device life and fault type prediction method and system based on digital twin technology | |
| CN114266289A (en) | Complex equipment health state assessment method | |
| CN111881574A (en) | Wind turbine generator key component reliability modeling method based on distribution function optimization | |
| CN115827411A (en) | Online monitoring and operation and maintenance evaluation system and method for automation equipment | |
| CN111680712B (en) | Method, device and system for predicting oil temperature of transformer based on similar time in day | |
| CN116539994A (en) | Substation main equipment operation state detection method based on multi-source time sequence data | |
| CN104866679A (en) | Mechanical characteristic prediction method of switching device | |
| CN115130764A (en) | Power distribution network situation prediction method and system based on state evaluation | |
| CN114548493A (en) | Method and system for predicting current overload of electric energy meter | |
| CN118607150B (en) | Edge optimization method and system for digital twin space modeling of power distribution network | |
| CN116432524A (en) | Transformer oil temperature prediction method, device, equipment and storage medium | |
| CN111311025A (en) | Load prediction method based on meteorological similar days | |
| CN116565850A (en) | Wind power ultra-short-term prediction method based on QR-BLSTM | |
| CN115081515A (en) | Energy efficiency evaluation model construction method and device, terminal and storage medium | |
| CN113392576A (en) | Staying balloon main cable tension state assessment early warning method | |
| CN118501692B (en) | Servo motor fault diagnosis method and system | |
| CN114565156B (en) | Power load prediction method, device, equipment and storage medium | |
| CN116150666B (en) | Energy storage system fault detection method and device and intelligent terminal | |
| CN118536410B (en) | Big data driven modeling-based energy consumption optimization decision analysis method and system | |
| WO2022059183A1 (en) | Information processing device, information processing method, and information processing program |
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 | ||
| GR01 | Patent grant |