Distribution computing power prediction method, system and medium based on digital twin

Abstract

The present application provides a distribution computing power prediction method, system and medium based on digital twins, and relates to the field of distribution prediction technology. The method includes: constructing an initial digital twin distribution model by collecting multi-dimensional power equipment information of the distribution system, and then preprocessing the real-time operation data obtained by dynamic monitoring, and then obtaining the target operation data, and then using the data to perform operation simulation to obtain simulation operation data, and then analyzing the simulation operation information to obtain the prediction result of the distribution computing power. The present application mainly solves the problem that the existing technology is not real-time enough, and the changes in the power grid state cannot be discovered and fed back in time, resulting in inaccurate prediction results. By monitoring and collecting the dynamic operation state and removing anomalies to obtain real-time data, and then simulating the distribution computing power through the digital twin distribution model, a more efficient and reliable energy forecast is achieved.

Description

Technical Field

[0001] The present invention relates to the field of power distribution prediction technology, and in particular to a power distribution computing power prediction method, system and medium based on digital twins. Background Art

[0002] With the rapid development and transformation of the power industry, the operation and control of power systems have become increasingly complex. In order to better meet this challenge, the distribution system needs more intelligent and efficient computing support to achieve more accurate predictions and more optimized control. Digital twin technology is a technology that connects the physical world with the digital world, which can realize real-time monitoring, simulation and prediction of the real world. In the distribution system, the application of digital twin technology can help us better understand the operating status and future trends of the system, so as to make more accurate predictions and decisions.

[0003] The existing technology is to train neural network models to learn the characteristics and patterns in historical data and predict future distribution computing power requirements.

[0004] The existing technology still has the problem of insufficient real-time performance, and changes in power grid status cannot be discovered and fed back in a timely manner, resulting in inaccurate prediction results. Summary of the invention

[0005] This application mainly solves the problem that the existing technology is not real-time enough and the changes in power grid status cannot be discovered and fed back in time, resulting in inaccurate prediction results.

[0006] In view of the above problems, the present application provides a distribution computing power prediction method, system and medium based on digital twins. In the first aspect, the embodiment of the present application provides a distribution computing power prediction method based on digital twins, and the method includes: constructing an initial digital twin distribution model through the multi-dimensional power equipment information collected from the distribution system. Preprocessing the real-time operation data of the distribution system obtained by dynamic monitoring to obtain target operation data. The initial digital twin distribution model uses the target operation data to simulate the operation of the distribution system to obtain simulated operation data. The distribution computing power prediction result of the distribution system is obtained based on the simulation operation information of the simulated operation data obtained by analysis.

[0007] In the second aspect, the present application provides a distribution computing power prediction system based on digital twins, the system comprising: an initial digital twin distribution model construction module, the initial digital twin distribution model construction module is used to construct an initial digital twin distribution model through the collected multi-dimensional power equipment information of the distribution system. A target operation data acquisition module, the target operation data acquisition module is used to pre-process the real-time operation data of the distribution system obtained by dynamic monitoring to obtain target operation data. A simulation operation data acquisition module, the simulation operation data acquisition module is the initial digital twin distribution model using the target operation data to simulate the operation of the distribution system to obtain simulation operation data. A prediction result acquisition module, the prediction result acquisition module is used to obtain the distribution computing power prediction result of the distribution system based on the simulation operation information of the simulation operation data obtained by analysis.

[0008] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed, the steps of any one of the methods described in the first aspect are implemented.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] The present application provides a distribution computing power prediction method, system and medium based on digital twins, and relates to the field of distribution prediction technology. The method includes: building an initial digital twin distribution model by collecting multi-dimensional power equipment information of the distribution system, and then preprocessing the real-time operation data obtained by dynamic monitoring, and then obtaining the target operation data, and then using the data to perform operation simulation to obtain simulated operation data, and then analyzing the simulated operation information to obtain the prediction result of the distribution computing power.

[0011] This application mainly solves the problem that the existing technology is not real-time enough, and the changes in the power grid status cannot be discovered and fed back in time, resulting in inaccurate prediction results. By monitoring and collecting the dynamic operating status and removing abnormalities to obtain real-time data, and then simulating the distribution computing power through the digital twin distribution model, a more efficient and reliable energy forecast is achieved.

[0012] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the present disclosure or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0014] Figure 1 A schematic flow chart of a power distribution computing power prediction method based on digital twins is provided for the embodiment of the present application;

[0015] Figure 2 A schematic flow chart of a method for acquiring real-time operation data in a power distribution computing power prediction method based on digital twins is provided for an embodiment of the present application;

[0016] Figure 3 A schematic flow chart of a method for acquiring target operation data in a power distribution computing power prediction method based on digital twins is provided for an embodiment of the present application;

[0017] Figure 4 A structural schematic diagram of a power distribution computing power prediction system based on digital twins is provided for an embodiment of the present application.

[0018] Explanation of the accompanying drawings: initial digital twin distribution model construction module 10, target operation data acquisition module 20, simulation operation data acquisition module 30, prediction result acquisition module 40. DETAILED DESCRIPTION

[0019] This application mainly solves the problem that the existing technology is not real-time enough, and the changes in the power grid status cannot be discovered and fed back in time, resulting in inaccurate prediction results. By monitoring and collecting the dynamic operating status and removing abnormalities to obtain real-time data, and then simulating the distribution computing power through the digital twin distribution model, a more efficient and reliable energy forecast is achieved.

[0020] In order to better understand the above technical solution, the above solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods:

[0021] Embodiment 1

[0022] like Figure 1 The power distribution computing power prediction method based on digital twin is shown, and the method includes:

[0023] Construct an initial digital twin distribution model through the collected multi-dimensional power equipment information of the distribution system;

[0024] Specifically, firstly, multi-dimensional power equipment information collection is carried out: detailed information of various equipment in the distribution system needs to be collected. Including equipment type, location, operating status, load condition, fault history, maintenance record, etc. This information can be obtained through various sensors, monitoring systems, power monitoring systems, etc. Then data cleaning and preprocessing are carried out: the collected raw data is cleaned and preprocessed to remove errors and abnormal data to ensure the quality and reliability of the data. This may include data format conversion, missing value processing, outlier detection, etc. Then model construction is carried out: three-dimensional modeling, virtual reality, and digital twins are used to build a digital twin distribution model. According to the collected power equipment information, a virtual model that is highly similar to the actual distribution system can be built, including the layout, connection relationship, operating status, etc. of the equipment. Data mapping and synchronization: the collected power equipment information is mapped to the digital twin distribution model and kept synchronized with the actual system. Including the collection, processing and updating of real-time data to ensure that the model can reflect the real-time status and changes of the actual system. Model verification and testing: the constructed digital twin distribution model is verified and tested to ensure its accuracy and reliability. The constructed digital twin distribution model is applied to the monitoring, control and optimization of the distribution system. The model can realize real-time monitoring of the system, predict future power demand and equipment status, perform intelligent scheduling and optimize operation, and improve the energy efficiency and reliability of the system.

[0025] Preprocessing the real-time operation data of the power distribution system obtained through dynamic monitoring to obtain target operation data;

[0026] Specifically, the real-time operation data of the distribution system is obtained through the dynamic monitoring system. These data may include current, voltage, power factor, harmonic content, etc. Sensors including current, voltage, power, and electricity, as well as data collectors and communication devices are used to collect data from the distribution system. According to the actual situation of the distribution system, select a suitable installation location, install sensors and other equipment into the distribution system, and ensure that the connection and function of the equipment and the system are normal. The real-time data of the distribution system is collected through the data collector, including current, voltage, power, electricity, etc., as well as parameters such as temperature and humidity of the environment and equipment. Then the collected raw data is cleaned to remove errors and abnormal data. Including data format conversion, missing value processing, outlier detection and correction, etc. The collected real-time data is converted and processed as necessary, including data normalization, standardization, smoothing, etc. The key data related to the operation status of the distribution system is screened out from the raw data. Including data related to equipment status, fault warning, load change, etc. Aggregate the screened key data to obtain more comprehensive target operation data. Including calculating aggregation indicators such as average value, cumulative amount, and change rate. The real-time operation data of the distribution system obtained through dynamic monitoring can be preprocessed to obtain more concise and effective target operation data, providing basic data support for subsequent distribution system operation analysis and optimization.

[0027] The initial digital twin power distribution model uses the target operation data to perform operation simulation of the power distribution system to obtain simulation operation data;

[0028] Specifically, data preparation: First, the collected target operation data needs to be prepared. Data cleaning, conversion, and aggregation are used to process the data so that it can be used in the digital twin distribution model. Model loading: The constructed initial digital twin distribution model is loaded into the simulation environment. This model includes the three-dimensional model, operation status, and connection relationship of power equipment and systems. Data mapping: Map the prepared target operation data to the digital twin distribution model. Input the data into the model and assign the characteristics and status of the data to the corresponding equipment or system. Simulation operation: Use the digital twin distribution model to perform simulation operation, including simulating the real-time operation of the distribution system, predicting future power demand and equipment status, etc. Data extraction: During the simulation operation, simulation operation data can be extracted. These data include the operation status of the equipment, the performance indicators of the system, energy consumption, etc. Result presentation: The results of the simulation operation are presented in the form of charts or reports to facilitate performance evaluation, decision making, etc. Digital twin technology and target operation data can be used to simulate the operation of the distribution system, so as to better understand the performance and behavior of the system and provide support for the optimization and control of the distribution system.

[0029] The power distribution computing power prediction result of the power distribution system is obtained based on the simulation operation information of the simulation operation data obtained by analysis.

[0030] Specifically, operation information extraction: extract the operation information of the simulation during the simulation operation. This information includes the operation status of the equipment, the performance indicators of the system, energy consumption, etc. Operation information analysis: conduct in-depth analysis of the extracted simulation operation information. This includes comparing the differences between the simulation results and the actual data, evaluating the performance indicators of the system, optimizing the operation strategy of the system, etc. At the same time, information related to the distribution computing power can be extracted, such as equipment load conditions, system topology, power quality, etc. Prediction model construction: according to the correlation between the extracted operation information and the distribution computing power, a prediction model is established using algorithms such as regression analysis, machine learning, and deep learning. Prediction result calculation: using the constructed prediction model, input the simulation operation information as input, and calculate the prediction results of the distribution computing power. This includes calculating the power demand of the system, equipment failure rate, energy loss, etc. In this way, the distribution computing power prediction results can be obtained based on the simulation operation information, and applied to the optimization and control of the distribution system to achieve more efficient and reliable energy management and operation.

[0031] Furthermore, the present application method also includes:

[0032] A power distribution computing power level that matches the power distribution computing power prediction result;

[0033] The power distribution system is dynamically regulated by starting a power distribution control scheme at the power distribution computing power level.

[0034] Specifically, distribution computing power level matching: According to the prediction results, the distribution computing power is divided into different levels. Including dividing the levels according to factors such as power demand, equipment load, and system performance indicators. Determine the computing power level target that needs to be matched, such as basic computing power, standard computing power, and advanced computing power. Distribution control plan formulation: Formulate corresponding distribution control plans for different distribution computing power levels. Including scheduling strategy adjustment, power load distribution, equipment maintenance plan, etc. Determine the distribution control plan: According to the actual situation and needs of the distribution system, formulate corresponding distribution control plans, including control strategies for adjusting voltage, frequency, power factor, etc. For example, at the basic computing power level, a rule-based scheduling strategy can be adopted, while at the advanced computing power level, an artificial intelligence-based scheduling strategy can be adopted. For different distribution computing power levels, the power load distribution plan can be adjusted. For example, at the basic computing power level, a strategy of average load distribution can be adopted, while at the advanced computing power level, a strategy of optimized load distribution can be adopted to achieve more efficient power load distribution and utilization. Start the power distribution control plan: Start the formulated power distribution control plan through software or hardware, connect it with the real-time data of the power distribution system, and realize the dynamic control of the power distribution system. Dynamic control implementation: According to the real-time operation of the power distribution system and the matching of the power distribution computing power level, select the corresponding power distribution control plan for implementation. Including automatic or manual adjustment of scheduling strategies, distribution of power loads, arrangement of equipment maintenance and other operations. According to the power distribution computing power level of the power distribution computing power prediction result, the corresponding power distribution control plan can be started to dynamically control the power distribution system, so as to realize the intelligent and optimized control of the power distribution system.

[0035] Furthermore, if Figure 2 As shown, the method of the present application, before preprocessing the real-time operation data of the power distribution system obtained by dynamic monitoring, includes:

[0036] Reading a predetermined computing power factor, wherein the predetermined computing power factor includes a meteorological factor, a load factor, and a fault factor;

[0037] Based on the meteorological factor monitoring, real-time meteorological data of the power distribution system is obtained;

[0038] Obtaining real-time power consumption data of the power distribution system based on the load factor monitoring;

[0039] Obtaining real-time equipment status data of the power distribution system based on the fault factor monitoring;

[0040] The real-time meteorological data, the real-time electricity consumption data and the real-time device status data together constitute the real-time operation data.

[0041] Specifically, read the predetermined computing power factor: read the predetermined computing power factor including meteorological factors, load factors and fault factors. These factors are pre-set or derived from historical data. Monitor the distribution system based on meteorological factors: obtain real-time meteorological data of the distribution system, such as temperature, humidity, wind speed, etc., through meteorological monitoring equipment or meteorological data interface. Monitor the distribution system based on load factors: obtain real-time power consumption data of the distribution system, such as current, voltage, power, etc., through power monitoring equipment or power data interface. These data reflect the real-time load of the distribution system. Monitor the distribution system based on fault factors: obtain real-time equipment status data of the distribution system, such as equipment failure, abnormal conditions, etc., through the equipment monitoring system or status monitoring interface. These data are used to evaluate the health of the system. Composition of real-time operation data: Real-time meteorological data, real-time power consumption data and real-time equipment status data are combined to form real-time operation data. These data provide comprehensive operation status information of the distribution system for distribution computing power prediction. Real-time operation data can be used as input and input into the distribution computing power prediction model based on digital twin technology. Use the prediction model to predict the distribution computing power and obtain the prediction result. According to the prediction results, the corresponding power distribution computing power level is matched, and the corresponding power distribution control plan is started for dynamic control. The control plan may include adjusting the dispatching strategy, allocating power load, maintaining equipment and other operations.

[0042] Furthermore, if Figure 3 As shown, the present application method also includes:

[0043] Performing index clustering on the data in the real-time operation data to obtain an index clustering result;

[0044] Extracting a first cluster from the indicator clustering result, where the first cluster refers to a clustering result of a first indicator corresponding to first abnormal data, and the first abnormal data refers to any one of the real-time operation data;

[0045] Performing an abnormal support analysis on the first abnormal data based on multiple data in the first cluster to obtain a first abnormal support;

[0046] If the first abnormal support reaches a predetermined threshold, the first abnormal data is removed from the real-time operation data to obtain the target operation data.

[0047] Specifically, data preprocessing: First, the collected real-time operation data is preprocessed, including data cleaning, missing value processing, etc., to ensure the quality and reliability of the data. Data clustering: Use clustering algorithms such as K-means and DBSCAN to cluster the preprocessed data. In the clustering process, it is necessary to consider the classification of indicators such as current, voltage, temperature, etc. to fully reflect the state of the distribution system. First clustering cluster extraction: After obtaining the indicator clustering results, select the first cluster, that is, the clustering result of the first indicator corresponding to the first abnormal data. The first abnormal data refers to any data in the real-time operation data, which may be abnormal or outlier. Abnormal support analysis: Based on multiple data in the first cluster, the first abnormal data is analyzed for abnormal support. This can be achieved by calculating the similarity or distance between the first abnormal data and other data in the first cluster. If the first abnormal data is significantly different from most of the data in the first cluster, it can be considered as abnormal data. Abnormal data removal: If the first abnormal support reaches the predetermined threshold, it means that the first abnormal data is really abnormal, so it is removed to avoid the erroneous data affecting the simulation accuracy and effectiveness of the digital twin model. Therefore, the first abnormal data is removed from the real-time operation data to obtain the target operation data. Abnormal data can be effectively detected and filtered, thereby improving the accuracy of distribution computing power prediction. At the same time, it can also reduce the impact of abnormal data on the regulation of the distribution system and improve the reliability and stability of the system.

[0048] Furthermore, in the method of the present application, the performing of abnormal support analysis on the first abnormal data based on the plurality of data in the first clustering clusters includes:

[0049] Read the exception support function;

[0050] The first abnormal data is subjected to support analysis by using the abnormal support function to obtain the first abnormal support, wherein the abnormal support function is expressed as follows:

[0051] sup(x1,x2)=k[1+(x1-x2) e ]-1;

[0052] Among them, sup(x1,x2) refers to the first anomaly support of any data x2 among the multiple data to the first anomaly data x1, k refers to the amplitude of the anomaly support function, and k∈[0,1], e refers to the attenuation factor, and e≥0.

[0053] Specifically, read the abnormal support function: read the abnormal support function from a preset function library or algorithm. This function will be used to analyze the support of the first abnormal data. Perform support analysis through the abnormal support function: take multiple data in the first cluster as input, and use the abnormal support function for calculation and analysis. According to the function definition, calculate the support of each data for the first abnormal data. Calculate the first abnormal support: according to the output of the abnormal support function, perform support analysis on the first abnormal data. Including calculating the mean, median, and standard deviation of the support.

[0054] Further, the method of the present application is applied to a distribution computing power prediction system based on digital twins, the system comprising a load simulation prediction unit and a fault simulation prediction unit, and the simulation operation information of the simulation operation data obtained by analysis is used to obtain the distribution computing power prediction result of the distribution system, including:

[0055] The load simulation prediction unit has a load prediction model embedded therein;

[0056] Using the target power consumption data in the target operation data as input information of the load forecasting model to obtain a target load forecast value;

[0057] Normalizing the target load prediction value to obtain a target load feedback adjustment coefficient;

[0058] The fault simulation prediction unit has a fault prediction model embedded in it;

[0059] Using the target device status data in the target operation data as input information of the fault prediction model to obtain a target fault prediction value;

[0060] Normalizing the target fault prediction value to obtain a target fault feedback adjustment coefficient;

[0061] A target meteorological feedback adjustment coefficient matching the target meteorological data in the target operation data;

[0062] The initial computing power of the power distribution system is adjusted according to the target load feedback adjustment coefficient, the target fault feedback adjustment coefficient and the target meteorological feedback adjustment coefficient to obtain the power distribution computing power prediction result.

[0063] Specifically, load simulation prediction: a load prediction model is embedded in the load simulation prediction unit, and the target power consumption data in the target operation data is used as the input information of the load prediction model. Through the load prediction model, the future power load situation can be predicted to obtain the target load prediction value. Load feedback regulation: the target load prediction value is normalized to obtain the target load feedback regulation coefficient. This coefficient reflects the accuracy of load prediction and can be used to adjust the operation strategy of the distribution system. Fault simulation prediction: a fault prediction model is embedded in the fault simulation prediction unit, and the target device status data in the target operation data is used as the input information of the fault prediction model. Through the fault prediction model, the future equipment failure situation can be predicted to obtain the target fault prediction value. Fault feedback regulation: the target fault prediction value is normalized to obtain the target fault feedback regulation coefficient. This coefficient reflects the accuracy of fault prediction and can be used to adjust the operation strategy of the distribution system. Meteorological feedback regulation: the target meteorological feedback regulation coefficient that matches the target meteorological data in the target operation data. This coefficient can adjust the operation strategy of the distribution system according to the historical data and future predictions of meteorological factors. According to the target load feedback adjustment coefficient, target fault feedback adjustment coefficient and target meteorological feedback adjustment coefficient, the initial computing power of the distribution system is adjusted to obtain the distribution computing power forecast result. These feedback adjustment coefficients can reflect the real-time status and future trends of the system and are used to adjust the computing power allocation and optimize the control strategy of the distribution system.

[0064] Embodiment 2

[0065] Based on the same inventive concept as the above-mentioned embodiment of a method for predicting power distribution computing power based on digital twins, as Figure 4 As shown, the present application provides a power distribution computing power prediction system based on digital twins, the system comprising:

[0066] An initial digital twin power distribution model construction module 10, wherein the initial digital twin power distribution model construction module 10 is used to construct an initial digital twin power distribution model through the collected multi-dimensional power equipment information of the power distribution system;

[0067] A target operation data acquisition module 20, wherein the target operation data acquisition module 20 is used to pre-process the real-time operation data of the power distribution system obtained by dynamic monitoring to obtain target operation data;

[0068] A simulation operation data acquisition module 30, wherein the simulation operation data acquisition module 30 is the initial digital twin distribution model using the target operation data to perform operation simulation of the distribution system to obtain simulation operation data;

[0069] The prediction result acquisition module 40 is used to obtain the distribution computing power prediction result of the distribution system according to the simulation operation information of the simulation operation data obtained by analysis.

[0070] Furthermore, the system also includes:

[0071] A power level matching module, used to match the power distribution computing power level of the power distribution computing power prediction result;

[0072] A dynamic control module is used to dynamically control the power distribution system by starting the power distribution control plan at the power distribution computing power level.

[0073] Furthermore, the system also includes:

[0074] A computing power factor reading module, used to read a predetermined computing power factor, wherein the predetermined computing power factor includes a meteorological factor, a load factor, and a fault factor;

[0075] A real-time meteorological data acquisition module is used to obtain real-time meteorological data of the power distribution system based on the meteorological factor monitoring;

[0076] A real-time power consumption data acquisition module is used to obtain the real-time power consumption data of the power distribution system based on the load factor monitoring;

[0077] A real-time device status data acquisition module is used to obtain the real-time device status data of the power distribution system based on the fault factor monitoring;

[0078] The real-time operation data acquisition module is a module that combines the real-time meteorological data, the real-time power consumption data and the real-time device status data to form the real-time operation data.

[0079] Furthermore, the system also includes:

[0080] A clustering result acquisition module is used to perform index clustering on the data in the real-time operation data to obtain an index clustering result;

[0081] A first clustering cluster extraction module, used to extract a first clustering cluster from the indicator clustering result, wherein the first clustering cluster refers to a clustering result of a first indicator corresponding to first abnormal data, and the first abnormal data refers to any data in the real-time operation data;

[0082] A first abnormal support acquisition module is configured to perform an abnormal support analysis on the first abnormal data based on a plurality of data in the first cluster to obtain a first abnormal support;

[0083] The target operation data acquisition module removes the first abnormal data from the real-time operation data to obtain the target operation data if the first abnormal support reaches a predetermined threshold.

[0084] Furthermore, the system also includes:

[0085] An abnormal support function reading module is used to read the abnormal support function;

[0086] A support function acquisition module is used to perform support analysis on the first abnormal data by using the abnormal support function to obtain the first abnormal support, wherein the abnormal support function is expressed as follows:

[0087] sup(x1,x2)=k[1+(x1-x2) e ]-1;

[0088] Among them, sup(x1,x2) refers to the first anomaly support of any data x2 among the multiple data to the first anomaly data x1, k refers to the amplitude of the anomaly support function, and k∈[0,1], e refers to the attenuation factor, and e≥0.

[0089] Furthermore, the system also includes:

[0090] A load prediction model acquisition module is a load prediction model embedded in the load simulation prediction unit;

[0091] A target load forecast value acquisition module, used to use the target power consumption data in the target operation data as input information of the load forecast model to obtain a target load forecast value;

[0092] A target load feedback adjustment coefficient acquisition module is used to normalize the target load prediction value to obtain the target load feedback adjustment coefficient;

[0093] A fault prediction model acquisition module is a fault prediction model embedded in the fault simulation prediction unit;

[0094] A target fault prediction value acquisition module, used to use the target device status data in the target operation data as input information of the fault prediction model to obtain a target fault prediction value;

[0095] A fault feedback adjustment coefficient acquisition module is used to normalize the target fault prediction value to obtain a target fault feedback adjustment coefficient;

[0096] A target meteorological feedback adjustment coefficient matching module, used to match the target meteorological feedback adjustment coefficient of the target meteorological data in the target operation data;

[0097] The power distribution computing power prediction result acquisition module is used to adjust the initial computing power of the power distribution system according to the target load feedback adjustment coefficient, the target fault feedback adjustment coefficient and the target meteorological feedback adjustment coefficient to obtain the power distribution computing power prediction result.

[0098] Embodiment 3

[0099] The present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed, the steps of any one of the methods described in Example 1 are implemented.

[0100] Through the detailed description of the distribution computing power prediction method based on digital twins mentioned above, those skilled in the art can clearly understand the distribution computing power prediction system based on digital twins in this embodiment. For the system disclosed in the embodiment, since it corresponds to the device disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0101] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A distribution computing power prediction method based on digital twins, characterized in that: include: Construct an initial digital twin distribution model through the collected multi-dimensional power equipment information of the distribution system; Preprocessing the real-time operation data of the power distribution system obtained through dynamic monitoring to obtain target operation data; The initial digital twin power distribution model uses the target operation data to perform operation simulation of the power distribution system to obtain simulation operation data; Obtaining a power distribution computing power prediction result of the power distribution system according to the simulation operation information of the simulation operation data obtained by analysis; Before preprocessing the real-time operation data of the power distribution system obtained by dynamic monitoring, the method includes: Reading a predetermined computing power factor, wherein the predetermined computing power factor includes a meteorological factor, a load factor, and a fault factor; Based on the meteorological factor monitoring, real-time meteorological data of the power distribution system is obtained; Obtaining real-time power consumption data of the power distribution system based on the load factor monitoring; Obtaining real-time equipment status data of the power distribution system based on the fault factor monitoring; The real-time meteorological data, the real-time power consumption data and the real-time device status data together constitute the real-time operation data; The method comprises: Performing index clustering on the data in the real-time operation data to obtain an index clustering result; Extracting a first cluster from the indicator clustering result, where the first cluster refers to a clustering result of a first indicator corresponding to first abnormal data, and the first abnormal data refers to any one of the real-time operation data; Performing an abnormal support analysis on the first abnormal data based on multiple data in the first cluster to obtain a first abnormal support; If the first abnormal support reaches a predetermined threshold, the first abnormal data is removed from the real-time operation data to obtain the target operation data.

2. The method according to claim 1, characterized in that: The method further comprises: A power distribution computing power level that matches the power distribution computing power prediction result; The power distribution system is dynamically regulated by starting a power distribution control scheme at the power distribution computing power level.

3. The method according to claim 1, characterized in that: The performing abnormal support analysis on the first abnormal data based on the plurality of data in the first clustering clusters includes: Read the exception support function; The first abnormal data is subjected to support analysis by using the abnormal support function to obtain the first abnormal support, wherein the abnormal support function is expressed as follows: ; in, Any data among the multiple data For the first abnormal data The first anomaly support, is the magnitude of the anomaly support function, and , is the attenuation factor, and .

4. The method according to claim 1, characterized in that: The method is applied to a power distribution computing power prediction system based on digital twins, the system comprising a load simulation prediction unit and a fault simulation prediction unit, and the power distribution computing power prediction result of the power distribution system is obtained according to the simulation operation information of the simulation operation data obtained by analysis, including: The load simulation prediction unit has a load prediction model embedded therein; Using the target power consumption data in the target operation data as input information of the load forecasting model to obtain a target load forecast value; Normalizing the target load prediction value to obtain a target load feedback adjustment coefficient; The fault simulation prediction unit has a fault prediction model embedded in it; Using the target device status data in the target operation data as input information of the fault prediction model to obtain a target fault prediction value; Normalizing the target fault prediction value to obtain a target fault feedback adjustment coefficient; A target meteorological feedback adjustment coefficient matching the target meteorological data in the target operation data; The initial computing power of the power distribution system is adjusted according to the target load feedback adjustment coefficient, the target fault feedback adjustment coefficient and the target meteorological feedback adjustment coefficient to obtain the power distribution computing power prediction result.

5. The power distribution computing power prediction system based on digital twins is characterized by: The system is used to execute the method according to any one of claims 1 to 4, comprising: An initial digital twin power distribution model construction module, wherein the initial digital twin power distribution model construction module is used to construct an initial digital twin power distribution model through the collected multi-dimensional power equipment information of the power distribution system; A target operation data acquisition module, the target operation data acquisition module is used to pre-process the real-time operation data of the power distribution system obtained by dynamic monitoring to obtain target operation data; A simulation operation data acquisition module, wherein the simulation operation data acquisition module is the initial digital twin distribution model using the target operation data to perform operation simulation of the distribution system to obtain simulation operation data; A prediction result acquisition module is used to obtain the distribution computing power prediction result of the distribution system based on the simulation operation information of the simulation operation data obtained by analysis.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which implements the steps of the method according to any one of claims 1 to 4 when executed.

Images