Distribution network transformation plan optimization method and system based on visualization and power outage parameters

Abstract

The present invention discloses a method and system for optimizing a distribution network transformation plan based on visualization and power outage parameters, and belongs to the technical field of distribution network construction. The method includes: mapping a scalable vector graphics file to be verified to a general information model to generate a general information model to be verified; extracting the basic topology of the distribution network, dividing the power equipment on the shortest path from the user node to the substation into necessary power equipment and non-necessary power equipment; constructing a LightGBM fault probability prediction model based on the impact of necessary power equipment and non-necessary power equipment on user power outages, and calculating the power outage parameters of the user node based on the model; calculating the power outage parameters of the substation system; and solving and obtaining the optimal distribution network transformation plan based on the power outage parameters of the user node and the substation system. The present invention improves the efficiency of the distribution network transformation plan, improves the reliability and real-time performance of the distribution network transformation, reduces manual intervention, and reduces the overall cost of the distribution network transformation.

Description

Technical Field

[0001] The present invention belongs to the technical field of distribution network construction, and more specifically, relates to a distribution network transformation plan optimization method and system based on visualization and power outage parameters. Background Art

[0002] The construction of distribution networks requires an analysis of their operational reliability. Based on the massive multi-source heterogeneous data provided by the operation and dispatching monitoring systems of the distribution networks, and taking into account the impact of factors such as the distribution system network structure and the parameters of each component, the real-time operating environment, the real-time operating status of the power grid, and the dispatching and maintenance plans on the system's operational reliability, the system's operational index values ​​are given in real time.

[0003] A key criterion for distribution network transformation is the grid's ability to continuously supply power—that is, the power system's ability to effectively provide electricity to users. This reflects the security and stability of the power supply system. Operational distribution network transformation emphasizes real-time performance, fully considering the potential impact of current equipment status and operating information on future stable operations. Existing technologies use historical data for distribution network transformation, ignoring real-time data on current equipment status and operating information in the real world. This results in a mismatch between historical equipment failure rates and current real-time data, leading to unstable distribution network transformations.

[0004] In the existing technology, it is relatively difficult to optimize the distribution network transformation plan. From changing the topology structure to quantifying the impact on users and substation systems is a long and complicated process. Each plan takes a long time from drawing to verification, and the workload of planners is heavy. Summary of the Invention

[0005] To address the deficiencies in the prior art, the present invention provides a distribution network transformation plan optimization method and system based on visualization and power outage parameters. By combining the distribution network topology information with the real-time equipment failure rate information, the LightGBM (Light Gradient Boosting Machine) is used to calculate the real-time power outage rate and power outage time indicators at the load point and even system level, and obtain indicators such as the real-time equivalent annual power outage number and average power outage duration of the distribution network system, thereby reflecting the feasibility of the current system transformation plan.

[0006] The present invention adopts the following technical solutions.

[0007] A first aspect of the present invention provides a method for optimizing a distribution network transformation plan based on visualization and power outage parameters, comprising the following steps:

[0008] Obtain a real-time scalable vector graphics file and a real-time universal information model of the distribution network to be transformed in the substation area, apply a transformation plan based on the scalable vector graphics file to generate a scalable vector graphics file to be verified, and map the scalable vector graphics file to be verified to the universal information model to generate a universal information model to be verified;

[0009] In the general information model to be verified, the basic topology of the distribution network is extracted, and the power equipment on the shortest path from the user node to the substation is divided into necessary power equipment and non-necessary power equipment;

[0010] Collect historical operation data of power equipment in the substation within a set time range, build a LightGBM fault probability prediction model based on the impact of necessary and non-essential power equipment on user power outages, and calculate the power outage parameters of user nodes based on the model, including: expected power outage rate and expected power outage duration;

[0011] Based on the power outage parameters of all user nodes in the substation, calculate the power outage parameters of the substation system, including the expected average power outage frequency and expected power outage duration of the substation;

[0012] Based on the power outage parameters of user nodes and the power outage parameters of the substation system, a multi-objective optimization model is constructed with the goal of minimizing the risk of user and system power outages. The optimal distribution network transformation plan is obtained, and the distribution network transformation plan optimization based on visualization and power outage parameters is realized.

[0013] Preferably, the step of obtaining a real-time scalable vector graphics file and a real-time universal information model of the distribution network to be transformed in the substation, applying a transformation plan on the basis of the scalable vector graphics file to generate a scalable vector graphics file to be verified, and mapping the scalable vector graphics file to be verified to the universal information model to generate the universal information model to be verified specifically includes:

[0014] Obtain a scalable vector graphics file of the current distribution network to be transformed in the substation area, and convert it into an editable mode of editable elements, preferably but not limited to, encoding the scalable vector graphics file into an extensible markup language format;

[0015] In editable mode, the transformation plan is applied to the distribution network to be transformed, including: adding and deleting power equipment and / or power lines to form the distribution network topology to be verified; after the compilation is completed, a scalable vector graphics file to be verified is generated;

[0016] The general information model of the current distribution network to be transformed is obtained, and the topological structure of the distribution network to be verified in the scalable vector graphics file to be verified is mapped to the general information model using a linked list structure to generate the general information model to be verified.

[0017] Preferably, the linked list structure includes the power equipment types, input end identification codes, output end identification codes, power equipment models and power line lengths added or deleted in the transformation plan.

[0018] Preferably, in the general information model to be verified, the basic topology of the distribution network is extracted, and the power equipment on the shortest path from the user node to the substation is divided into necessary power equipment and non-necessary power equipment, specifically including:

[0019] Extracting the basic topology structure of the distribution network from the general information model to be verified, and obtaining a basic topology structure diagram of the distribution network to be verified;

[0020] Based on the basic topology diagram of the distribution network to be verified, according to the distance between the user node and the substation, the shortest path for each user to reach the substation is traversed and searched. According to all the shortest paths, all the power equipment in the basic topology diagram of the distribution network are divided into necessary power equipment and non-necessary power equipment. The power equipment that appears in any shortest path is the necessary power equipment, and the power equipment that does not appear in any shortest path is the non-necessary power equipment.

[0021] Preferably, the historical operation data of the power equipment in the substation within the set time range is collected, and a LightGBM fault probability prediction model is constructed based on the impact of the necessary power equipment and non-essential power equipment on the user's power outage. The power outage parameters of the user node are calculated based on the model, specifically including:

[0022] Collect historical operation data of power equipment in the substation within the set time range and perform pre-processing;

[0023] Based on the preprocessed data set, train the LightGBM fault probability prediction model to obtain the trained LightGBM model;

[0024] The real-time equipment status and environmental factors are input into the trained LightGBM model, and the posterior probability is dynamically updated through the Markov Chain Monte Carlo method. The equipment failure probability matrix is ​​obtained based on the necessary power equipment and non-necessary power equipment.

[0025] Monte Carlo simulation is performed based on the equipment failure probability matrix to calculate the expected power outage rate and expected power outage duration of the user node.

[0026] Preferably, the method of training the LightGBM fault probability prediction model based on the preprocessed data set to obtain a trained LightGBM model specifically includes:

[0027] Based on the logarithmic loss function, with the goal of minimizing the difference between the predicted fault probability and the actual label, the objective function of the LightGBM fault probability prediction model is constructed, and the loss function is solved;

[0028] The device sample state feature values ​​are used to optimize the split nodes and leaf weights of each decision tree using the gradient descent method to generate a decision tree.

[0029] Calculate the weight of each tree by combining the decision tree with the loss function and back propagation;

[0030] When the validation set loss function does not decrease after K consecutive rounds, the training is terminated, the trained LightGBM model is obtained, and the optimal tree set and optimal weights are saved.

[0031] Preferably, the weight of each tree is calculated by combining the decision tree with the loss function and back propagation, specifically including:

[0032] For each sample i, calculate the prediction error;

[0033] Combined with the loss value, multiply the prediction error by the output value of the t-th tree, accumulate all sample results, and obtain the weight gradient of the loss function with respect to the t-th tree;

[0034] Calculate the weight of each tree based on the gradient.

[0035] Preferably, the real-time device status and environmental factors are input into the trained LightGBM model, the posterior probability is dynamically updated by the Markov Chain Monte Carlo method, and the device failure probability matrix is ​​obtained based on the necessary power equipment and non-necessary power equipment, specifically including:

[0036] Set the Bayesian prior for the optimal weight of each tree and obtain the initialized posterior distribution;

[0037] Based on the initialized posterior distribution, real-time device status, and environmental factors, the weight parameters are updated using the Metropolis-Hastings algorithm.

[0038] Calculate the acceptance probability based on the candidate parameters, compare the joint probability of the candidate parameters and the current parameters, determine whether to accept the new parameters, and set the weight obtained from the judgment result as the updated weight;

[0039] According to the updated weights, the predicted failure probability is solved and set as the basic failure probability. Combining the necessary power equipment and non-necessary power equipment, the failure probability of the equipment is filled in the corresponding position with the equipment ID as the row and the equipment status as the column to generate a failure probability matrix.

[0040] Preferably, performing a Monte Carlo simulation based on the device failure probability matrix to calculate the expected power outage rate and the expected power outage duration of the user node specifically includes:

[0041] Independently sample each device j and generate a random number based on the fault probability value corresponding to the device state. If the number is less than the fault probability, it is considered a fault.

[0042] Count the failure probabilities of all devices, weight them by the inverse of the number of devices on the path, and get the expected power outage rate of the user node;

[0043] The repair time of all faulty devices is counted and weighted by the inverse number of devices on the path to obtain the expected power outage duration of the user node.

[0044] A second aspect of the present invention provides a distribution network transformation plan optimization system based on visualization and power outage parameters, which runs the distribution network transformation plan optimization method based on visualization and power outage parameters described in the first aspect, including:

[0045] The data acquisition module is used to obtain the scalable vector graphics file and general information model of the distribution network to be transformed in the substation area, as well as the historical operation data of the power equipment in the substation area within a set time range;

[0046] A visual data conversion module is used to apply a transformation scheme on the basis of the scalable vector graphics file to generate a scalable vector graphics file to be verified, and map the scalable vector graphics file to be verified to a general information model to generate a general information model to be verified;

[0047] The equipment classification module is used to extract the basic topology of the distribution network from the general information model to be verified, and classify the power equipment on the shortest path from the user node to the substation into necessary power equipment and non-necessary power equipment;

[0048] The user node power outage parameter solution module is used to collect historical operation data of power equipment in the substation within a set time range, build a LightGBM fault probability prediction model based on the impact of necessary power equipment and non-essential power equipment on user power outages, and calculate the power outage parameters of the user node based on the model, including: expected power outage rate and expected power outage duration;

[0049] The power outage parameter calculation module of the substation system is used to calculate the power outage parameters of the substation system based on the power outage parameters of all user nodes in the substation, including the expected average power outage frequency and expected power outage duration of the substation;

[0050] The output module is used to build a multi-objective optimization model based on the power outage parameters of the user node and the power outage parameters of the substation system, with the goal of minimizing the risk of power outages for users and the system, to solve the optimal distribution network transformation plan, and realize the optimization of the distribution network transformation plan based on visualization and power outage parameters.

[0051] Compared with the prior art, the beneficial effects of the present invention include at least:

[0052] Generate transformation plans through visual editing of vector graphics files, automatically map them to a common information model, eliminate programming steps, improve efficiency, build a multi-objective optimization model, automatically quantify the impact of the plan on user and system outage parameters, replace manual trial and error, simplify the transformation plan design process, and improve the efficiency of distribution network transformation plans;

[0053] Since a failure of a necessary device on the main feeder will cause a power outage for all users along the path, a failure of a non-necessary device will only affect a certain area. Based on the shortest path, necessary devices and non-necessary devices are divided. The key risks are amplified by the necessary device weight correction factor, and the success probability of the isolation device introduced by non-necessary devices is used to quantify the effectiveness of the protection device, improve the accuracy of fault prediction, and reduce the calculation error of user-level power outage parameters.

[0054] Based on the LightGBM model and real-time data, the fault probability is dynamically predicted, combined with the Markov chain Monte Carlo to update the posterior probability, and the latest data is integrated in real time to improve the reliability of the model's posterior probability and the accuracy of fault prediction. The power outage parameters are calculated in real time through Monte Carlo simulation, which can dynamically reflect the impact of the current equipment status, adapt to environmental changes, and improve the stability, reliability and real-time performance of the distribution network transformation;

[0055] The multi-objective model automatically selects the optimal solution, balances the risks of user-level and system-level power outages, avoids excessively redundant designs, reduces manual intervention, and lowers the overall cost of distribution network transformation. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 This is a flow chart of a method for optimizing a distribution network transformation plan based on visualization and real-time power outage parameters according to an embodiment of the present invention;

[0057] Figure 2 It is a schematic diagram of extracting the basic topology structure of the distribution network in the general information model to be verified according to an embodiment of the present invention. DETAILED DESCRIPTION

[0058] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. The described embodiments are only part of the embodiments of the present invention, not all of them. Based on the spirit of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of the present invention.

[0059] like Figure 1 As shown, embodiment 1 of the present invention provides a method for optimizing a distribution network transformation plan based on visualization and real-time power outage parameters, comprising the following steps:

[0060] Step 1: Obtain the scalable vector graphics file of the distribution network to be transformed in the substation, apply the transformation plan based on the scalable vector graphics file, generate the scalable vector graphics file to be verified, and map the scalable vector graphics file to be verified to the general information model to generate the general information model to be verified.

[0061] In a preferred but non-limiting embodiment of the present invention, step 1 specifically comprises:

[0062] Step 1.1: Obtain a scalable vector graphics file of the current distribution network to be transformed in the substation area, and convert it into an editable mode of editable elements. Preferably, but not limited to, encode the scalable vector graphics file into an extensible markup language format.

[0063] Step 1.2: In editable mode, apply the transformation plan to the distribution network to be transformed, including: adding and deleting power equipment and / or power lines to form the distribution network topology to be verified; after the compilation is completed, generate a scalable vector graphics file to be verified.

[0064] Step 1.3: Obtain the general information model of the current distribution network to be transformed, map the distribution network topology to be verified in the scalable vector graphics file to be verified to the general information model using a linked list structure, and generate the general information model to be verified.

[0065] Preferably, the linked list structure includes parameters such as the power equipment type, input end identification code, output end identification code, power equipment model, and power line length that are added or deleted in the transformation plan.

[0066] It is worth noting that the addition and deletion of power equipment and / or power lines is based on scalable vector graphics files rather than general information models, and can be carried out in a visual manner without having to enter them in the form of programming language codes, thereby improving the work efficiency of planning and design personnel.

[0067] Step 2: In the general information model to be verified, extract the basic topology of the distribution network and divide the power equipment on the shortest path from the user node to the substation into necessary power equipment and non-necessary power equipment. Figure 2 shown.

[0068] In a preferred but non-limiting embodiment of the present invention, step 2 specifically comprises:

[0069] Step 2.1: Extract the basic topology of the distribution network from the general information model to be verified, including busbars, circuit breakers, disconnectors, distribution transformers, overhead lines, and cables, to obtain a basic topology diagram of the distribution network to be verified.

[0070] The basic topology of the extraction distribution network specifically includes:

[0071] Merge elements: The adjacent cables and overhead lines of the same model will be set between nodes, including: between the starting point and the branch node, between the branch nodes, and between the branch node and the end point.

[0072] Ignored elements: Ignore electrical equipment such as oscilloscopes, current transformers and lightning arresters.

[0073] Understandably, the complete general information model for power systems is very large, encompassing a series of power system models, including core packages, topology packages, and protection packages. DL / T 890.301-2016 provides detailed specifications. During distribution network transformation verification, it's not necessary to load the entire model. The present invention extracts only key information, which is then mapped to the general information model using scalable vector graphics files, enabling verification with minimal effort.

[0074] Step 2.2: Based on the basic topology diagram of the distribution network to be verified, start traversing and searching for the shortest path for each user to reach the substation according to the distance between the user node, that is, the load node, and the substation. According to all the shortest paths, all the power equipment in the basic topology diagram of the distribution network are divided into necessary power equipment and non-necessary power equipment. The power equipment that appears in any shortest path is the necessary power equipment, and the power equipment that does not appear in any shortest path is the non-necessary power equipment.

[0075] Without taking backup power sources into account, the outage of essential power equipment will inevitably cause a power outage for users along the shortest path to which it is located; the outage of non-essential power equipment will not necessarily cause a power outage for users downstream of it. Furthermore, if the non-essential power equipment is located on a branch line and a fault isolation device is installed upstream of it, the outage of the non-essential power equipment will not cause a power outage for other branches. If the non-essential power equipment is located on a main feeder line, the outage of the non-essential power equipment downstream of the disconnector or sectionalizer will only cause a power outage for users upstream of it for the operating time of the disconnector or sectionalizer.

[0076] Step 3: Collect historical operation data of power equipment in the substation within the set time range, build a LightGBM fault probability prediction model based on the impact of necessary power equipment and non-essential power equipment on user power outages, and calculate the power outage parameters of the user node based on the model, including: expected power outage rate and expected power outage duration.

[0077] In a preferred but non-limiting embodiment of the present invention, step 3 specifically comprises:

[0078] Step 3.1: Collect historical operation data of power equipment in the substation within a set time range and perform preprocessing. The historical operation data of power equipment includes historical status sequence of equipment, environmental data and fault records.

[0079] Further preferably, step 3.1 includes:

[0080] Step 3.1.1: Collect historical operation data of power equipment in the substation within the set time range.

[0081] More preferably, step 3.1.1 comprises:

[0082] Collect historical equipment status sequences, including the status of power equipment operating normally, the status of power equipment requiring attention, the status of power equipment operating abnormally, and the status of power equipment with major abnormalities;

[0083] Collect environmental data, meteorological conditions, temperature, humidity and other environmental parameters;

[0084] Collect fault records of power equipment. The time range is preferably but not limited to the most recent year.

[0085] Step 3.1.2: Preprocess the historical operation data of the power equipment.

[0086] More preferably, step 3.1.2 comprises:

[0087] Convert the equipment history status sequence into numerical labels and add binary features based on the fault records of the power equipment S 关键性 , identifies whether the device is a mandatory device, if it is 1, otherwise it is 0, and obtains the device sample status characteristic value S=[S 1 ,S 2 ,S 3 , S 4 ,S 关键性 ] ,in, S 1 indicates the normal operation of the power equipment. S 2 indicates the status of the power equipment that needs attention. S 3 indicates abnormal operation of power equipment status, S 4 indicates a major abnormality in the power equipment status;

[0088] Normalize the meteorological data according to the disaster level to obtain the environmental factor ;

[0089] Combine the equipment sample state eigenvalues ​​and environmental factors to obtain the preprocessed data set, including the feature matrix .

[0090] Step 3.2: Based on the preprocessed data set, train the LightGBM fault probability prediction model to obtain the trained LightGBM model, which includes the tree set {ht} and weight {βt}.

[0091] Further preferably, step 3.2 includes:

[0092] In step 3.2.1, based on the logarithmic loss function, with the goal of minimizing the difference between the predicted fault probability and the actual label, the objective function of the LightGBM fault probability prediction model is constructed, and the loss value is solved, which is expressed as the following formula:

[0093]

[0094] Where,

[0095] is the loss value,

[0096] is the total number of training samples;

[0097] is the true label of the i-th sample, 0 means no fault, 1 means fault;

[0098] is the predicted failure probability of the i-th sample.

[0099] In step 3.2.2, the device sample state feature values ​​are used to optimize the split nodes and leaf weights of each decision tree using the gradient descent method to generate a decision tree.

[0100] Step 3.2.3, calculate the weight of each tree by combining the decision tree with the loss function and back propagation, specifically including:

[0101] For each sample i, calculate the prediction error ;

[0102] Combine the loss value and the prediction error with the output value of the tth tree Multiply and accumulate all sample results to get the gradient , the weight of each tree is calculated according to the gradient, which is expressed as follows:

[0103]

[0104] Where,

[0105] Represents the loss function For the first Tree weight gradient;

[0106] represents the true label of the i-th sample, 0 means no fault, 1 means fault;

[0107] Indicates the The output value of the decision tree for sample i;

[0108] represents the device status of the i-th sample;

[0109] represents the environmental factor of the i-th sample;

[0110] represents the predicted failure probability of the i-th sample, which is expressed as follows:

[0111]

[0112] Where,

[0113] Represents the Sigmoid function;

[0114] Indicates the number of decision trees;

[0115] represents the weight of the t-th tree in the k-th iteration;

[0116] Represents the learning rate, which is used to control the step size of weight update.

[0117] Step 3.2.4: When the validation set loss function does not decrease for K consecutive rounds, terminate the training, obtain the trained LightGBM model, and save the optimal tree set and optimal weights.

[0118] In step 4, the real-time equipment status and environmental factors are input into the trained LightGBM model, the posterior probability is dynamically updated through the Markov Chain Monte Carlo method, and the equipment failure probability matrix is ​​obtained based on the necessary power equipment and non-necessary power equipment.

[0119] In a preferred but non-limiting embodiment of the present invention, step 4 comprises:

[0120] Step 4.1: Set the Bayesian prior for the optimal weight of each tree and obtain the initialized posterior distribution, which is expressed as the following formula:

[0121]

[0122] Where,

[0123] is the initialized posterior distribution,

[0124] is a normal distribution,

[0125] As the mean, the initial weight obtained by training Set to mean,

[0126] is the variance.

[0127] Step 4.2: Update the weight parameters using the Metropolis-Hastings algorithm based on the initialized posterior distribution, real-time device status, and environmental factors. This includes:

[0128] The proposal distribution is generated with the current weight as the center, and candidate parameters are generated according to the variance, which can be expressed as the following formula:

[0129]

[0130] Where,

[0131] is a candidate parameter;

[0132] is the variance of the proposed distribution.

[0133] In step 4.3, the acceptance probability is calculated based on the candidate parameters. The joint probability of the candidate parameters and the current parameters is compared to determine whether to accept the new parameters. The weight obtained from the judgment result is set as the updated weight, which can be expressed as the following formula:

[0134]

[0135] Where,

[0136] is the probability of acceptance,

[0137] Is the likelihood function, indicating that at the current weight The probability of observing new data.

[0138] like ,accept , otherwise keep ,in, Represents a uniform distribution.

[0139] In step 4.4, based on the updated weights, the predicted failure probability is solved according to formula 4 and set as the basic failure probability. Combine the necessary power equipment and non-necessary power equipment, with the equipment ID as the row and the equipment status as the column, fill in the failure probability of the equipment in the corresponding position to generate a failure probability matrix, which is expressed as the following formula:

[0140]

[0141] Where,

[0142] Indicates the probability of failure of the necessary equipment;

[0143] Indicates the failure probability of non-essential equipment;

[0144] represents the basic failure probability;

[0145] Indicates the necessary equipment weight correction factor;

[0146] Represents the success probability of the isolation device.

[0147] Step 5: Perform Monte Carlo simulation based on the fault probability matrix to calculate the power outage parameters of the user nodes. Based on the power outage parameters of all user nodes in the substation, calculate the power outage parameters of the substation system, including the expected average power outage frequency and expected power outage duration of the substation.

[0148] In a preferred but non-limiting embodiment of the present invention, step 5 comprises:

[0149] In step 5.1, perform independent sampling on each device j and generate a random number based on the fault probability value corresponding to the device state. If the value is less than the fault probability, it is considered a fault. The expected power outage rate and the expected power outage duration of the user node are solved and expressed as the following formula:

[0150]

[0151] Where,

[0152] For user nodes The expected power outage rate,

[0153] For user nodes The set of devices on the shortest path to the substation;

[0154] is the failure probability of device j;

[0155] For user nodes Expected duration of power outage;

[0156] For equipment Historical repair time.

[0157] Step 5.2: Calculate the expected average outage frequency and outage duration in the area based on the users, expressed as follows:

[0158]

[0159] Where:

[0160] is the expected average power outage frequency in the substation area;

[0161] The expected duration of power outage in the substation;

[0162] is the total number of user nodes in the area, .

[0163] Step 6: Based on the outage parameters of the user nodes and the outage parameters of the substation system, a multi-objective optimization model is constructed with the goal of minimizing the risk of user and system outages. The optimal distribution network transformation plan is obtained, and the optimization of the distribution network transformation plan based on visualization and outage parameters is achieved, which is expressed as the following formula:

[0164]

[0165] Where:

[0166] The objective function of the transformation plan is to achieve a comprehensive reduction in power outage parameters for users and substations;

[0167] User node before transformation Expected power outage rate;

[0168] User node before transformation Expected duration of power outage;

[0169] is a constraint condition;

[0170] are the first and second power outage rate weights,

[0171] are the weights of the first and second power outage times.

[0172] Embodiment 2 of the present invention provides a distribution network transformation plan optimization system based on visualization and power outage parameters, which runs the distribution network transformation plan optimization method based on visualization and power outage parameters described in embodiment 1, including:

[0173] The data acquisition module is used to obtain the scalable vector graphics file and general information model of the distribution network to be transformed in the substation area, as well as the historical operation data of the power equipment in the substation area within a set time range;

[0174] A visual data conversion module is used to apply a transformation scheme on the basis of the scalable vector graphics file to generate a scalable vector graphics file to be verified, and map the scalable vector graphics file to be verified to a general information model to generate a general information model to be verified;

[0175] The equipment classification module is used to extract the basic topology of the distribution network from the general information model to be verified, and classify the power equipment on the shortest path from the user node to the substation into necessary power equipment and non-necessary power equipment;

[0176] The user node power outage parameter solution module is used to collect historical operation data of power equipment in the substation within a set time range, build a LightGBM fault probability prediction model based on the impact of necessary power equipment and non-essential power equipment on user power outages, and calculate the power outage parameters of the user node based on the model, including: expected power outage rate and expected power outage duration;

[0177] The power outage parameter calculation module of the substation system is used to calculate the power outage parameters of the substation system based on the power outage parameters of all user nodes in the substation, including the expected average power outage frequency and expected power outage duration of the substation;

[0178] The output module is used to build a multi-objective optimization model based on the power outage parameters of the user node and the power outage parameters of the substation system, with the goal of minimizing the risk of power outages for users and the system, to solve the optimal distribution network transformation plan, and realize the optimization of the distribution network transformation plan based on visualization and power outage parameters.

[0179] Compared with the prior art, the beneficial effects of the present invention include at least:

[0180] Generate transformation plans through visual editing of vector graphics files, automatically map them to a common information model, eliminate programming steps, improve efficiency, build a multi-objective optimization model, automatically quantify the impact of the plan on user and system outage parameters, replace manual trial and error, simplify the transformation plan design process, and improve the efficiency of distribution network transformation plans;

[0181] Since a failure of a necessary device on the main feeder will cause a power outage for all users along the path, a failure of a non-necessary device will only affect a certain area. Based on the shortest path, necessary devices and non-necessary devices are divided. The key risks are amplified by the necessary device weight correction factor, and the success probability of the isolation device introduced by non-necessary devices is used to quantify the effectiveness of the protection device, improve the accuracy of fault prediction, and reduce the calculation error of user-level power outage parameters.

[0182] Based on the LightGBM model and real-time data, the fault probability is dynamically predicted. In combination with the Markov chain Monte Carlo to update the posterior probability, the latest data is integrated in real time to improve the reliability of the model's posterior probability and the accuracy of fault prediction. The power outage parameters are calculated in real time through Monte Carlo simulation, which can dynamically reflect the impact of the current equipment status, adapt to environmental changes, and improve the reliability and real-time performance of the distribution network transformation.

[0183] The multi-objective model automatically selects the optimal solution, balances the risks of user-level and system-level power outages, avoids excessively redundant designs, reduces manual intervention, and lowers the overall cost of distribution network transformation.

[0184] The present disclosure may be a system, method and / or computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0185] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A method for optimizing distribution network transformation schemes based on visualization and power outage parameters, characterized in that: The following steps are involved: Obtain a real-time scalable vector graphics file and a real-time universal information model of the distribution network to be transformed in the substation area, apply a transformation plan based on the scalable vector graphics file to generate a scalable vector graphics file to be verified, and map the scalable vector graphics file to be verified to the universal information model to generate a universal information model to be verified; Extracting a basic topology structure of the distribution network from the general information model to be verified, and dividing the power equipment on the shortest path from the user node to the substation into necessary power equipment and non-necessary power equipment, including: extracting the basic topology structure of the distribution network from the general information model to be verified, and obtaining a basic topology structure diagram of the distribution network to be verified; Based on the basic topology diagram of the distribution network to be verified, according to the distance between the user node and the substation, the shortest path from each user to the substation is traversed and searched. According to all the shortest paths, all the power equipment in the basic topology diagram of the distribution network are divided into necessary power equipment and non-necessary power equipment. The power equipment that appears in any shortest path is set as the necessary power equipment, and the power equipment that does not appear in any shortest path is set as the non-necessary power equipment. Collect historical operation data of power equipment in the substation within a set time range, and build a LightGBM fault probability prediction model based on the impact of power outages caused by necessary and non-essential power equipment on users; Input the real-time device status and environmental factors into the trained LightGBM model to obtain the device failure probability matrix; Calculate the power outage parameters of the user node based on the equipment failure probability matrix, and calculate the power outage parameters of the substation system based on the power outage parameters of all user nodes in the substation. The power outage parameters of the user node include the expected power outage rate and the expected power outage duration, and the power outage parameters of the substation system include the expected average power outage frequency and the expected power outage duration. Based on the power outage parameters of user nodes and the power outage parameters of the substation system, a multi-objective optimization model is constructed with the goal of minimizing the risk of user and system power outages. The optimal distribution network transformation plan is obtained, and the distribution network transformation plan optimization based on visualization and power outage parameters is realized.

2. The method for optimizing distribution network transformation schemes based on visualization and power outage parameters according to claim 1, characterized in that: The method of obtaining a real-time scalable vector graphics file and a real-time universal information model of the distribution network to be transformed in the substation, applying a transformation plan based on the scalable vector graphics file to generate a scalable vector graphics file to be verified, and mapping the scalable vector graphics file to be verified to the universal information model to generate the universal information model to be verified specifically includes: Obtain a scalable vector graphics file of the current distribution network to be transformed in the substation area and convert it into an editable mode with editable elements; In editable mode, the transformation plan is applied to the distribution network to be transformed, including: adding and deleting power equipment and / or power lines to form the distribution network topology to be verified; after the compilation is completed, a scalable vector graphics file to be verified is generated; The general information model of the current distribution network to be transformed is obtained, and the topological structure of the distribution network to be verified in the scalable vector graphics file to be verified is mapped to the general information model using a linked list structure to generate the general information model to be verified.

3. The method for optimizing distribution network transformation schemes based on visualization and power outage parameters according to claim 1, characterized in that: The method collects historical operation data of power equipment in the substation within a set time range, and builds a LightGBM fault probability prediction model based on the impact of power outages caused by necessary power equipment and non-essential power equipment on users, specifically including: Collect historical operation data of power equipment in the substation within the set time range and perform pre-processing; Based on the preprocessed data set, the LightGBM fault probability prediction model is trained to obtain the trained LightGBM model.

4. The method for optimizing distribution network transformation schemes based on visualization and power outage parameters according to claim 3 is characterized in that: The method of training the LightGBM fault probability prediction model based on the preprocessed data set to obtain the trained LightGBM model specifically includes: Based on the logarithmic loss function, with the goal of minimizing the difference between the predicted fault probability and the actual label, the objective function of the LightGBM fault probability prediction model is constructed, and the loss function is solved; The device sample state feature values ​​are used to optimize the split nodes and leaf weights of each decision tree using the gradient descent method to generate a decision tree. Calculate the weight of each tree by combining the decision tree with the loss function and back propagation; When the validation set loss function does not decrease after K consecutive rounds, the training is terminated, the trained LightGBM model is obtained, and the optimal tree set and optimal weights are saved.

5. The method for optimizing distribution network transformation schemes based on visualization and power outage parameters according to claim 4, characterized in that: The weight of each tree is calculated by combining the decision tree with the loss function and back propagation, including: For each sample i, calculate the prediction error; Combined with the loss value, multiply the prediction error by the output value of the t-th tree, accumulate all sample results, and obtain the weight gradient of the loss function with respect to the t-th tree; Calculate the weight of each tree based on the gradient.

6. The method for optimizing distribution network transformation schemes based on visualization and power outage parameters according to claim 5, characterized in that: Input the real-time device status and environmental factors into the trained LightGBM model to obtain the device failure probability matrix, which specifically includes: The real-time equipment status and environmental factors are input into the trained LightGBM model, and the posterior probability is dynamically updated through the Markov Chain Monte Carlo method. The equipment failure probability matrix is ​​obtained based on the necessary power equipment and non-necessary power equipment.

7. The method for optimizing distribution network transformation schemes based on visualization and power outage parameters according to claim 6, characterized in that: The real-time device status and environmental factors are input into the trained LightGBM model, and the posterior probability is dynamically updated through the Markov Chain Monte Carlo method. The device failure probability matrix is ​​obtained based on the necessary power equipment and non-necessary power equipment, specifically including: Set the Bayesian prior for the optimal weight of each tree and obtain the initialized posterior distribution; Based on the initialized posterior distribution, real-time device status, and environmental factors, the weight parameters are updated using the Metropolis-Hastings algorithm. Calculate the acceptance probability based on the candidate parameters, compare the joint probability of the candidate parameters and the current parameters, determine whether to accept the new parameters, and set the weight obtained from the judgment result as the updated weight; According to the updated weights, the predicted failure probability is solved and set as the basic failure probability. Combining the necessary power equipment and non-necessary power equipment, the failure probability of the equipment is filled in the corresponding position with the equipment ID as the row and the equipment status as the column to generate a failure probability matrix.

8. The method for optimizing distribution network transformation schemes based on visualization and power outage parameters according to claim 6, characterized in that: The calculation of the power outage parameters of the user node based on the device failure probability matrix specifically includes: Independently sample each device j and generate a random number based on the fault probability value corresponding to the device state. If the number is less than the fault probability, it is considered a fault. Count the failure probabilities of all devices, weight them by the inverse of the number of devices on the path, and get the expected power outage rate of the user node; The repair time of all faulty devices is counted and weighted by the inverse number of devices on the path to obtain the expected power outage duration of the user node.

9. A distribution network transformation plan optimization system based on visualization and power outage parameters, which runs a distribution network transformation plan optimization method based on visualization and power outage parameters according to any one of claims 1 to 8, characterized in that: The data acquisition module is used to obtain the scalable vector graphics file and general information model of the distribution network to be transformed in the substation area, as well as the historical operation data of the power equipment in the substation area within a set time range; A visual data conversion module is used to apply a transformation scheme on the basis of the scalable vector graphics file to generate a scalable vector graphics file to be verified, and map the scalable vector graphics file to be verified to a general information model to generate a general information model to be verified; The equipment division module is used to extract the basic topology of the distribution network from the general information model to be verified, and divide the power equipment on the shortest path from the user node to the substation into necessary power equipment and non-necessary power equipment, including: extracting the basic topology of the distribution network from the general information model to be verified to obtain a basic topology diagram of the distribution network to be verified; Based on the basic topology diagram of the distribution network to be verified, according to the distance between the user node and the substation, the shortest path from each user to the substation is traversed and searched. According to all the shortest paths, all the power equipment in the basic topology diagram of the distribution network are divided into necessary power equipment and non-necessary power equipment. The power equipment that appears in any shortest path is set as the necessary power equipment, and the power equipment that does not appear in any shortest path is set as the non-necessary power equipment. The user node power outage parameter solution module is used to collect historical operation data of power equipment in the substation within a set time range, build a LightGBM fault probability prediction model based on the impact of necessary power equipment and non-essential power equipment on user power outages, and calculate the power outage parameters of the user node based on the model, including: expected power outage rate and expected power outage duration; The power outage parameter calculation module of the substation system is used to calculate the power outage parameters of the substation system based on the power outage parameters of all user nodes in the substation, including the expected average power outage frequency and expected power outage duration of the substation; The output module is used to build a multi-objective optimization model based on the power outage parameters of the user node and the power outage parameters of the substation system, with the goal of minimizing the risk of power outages for users and the system, to solve the optimal distribution network transformation plan, and realize the optimization of the distribution network transformation plan based on visualization and power outage parameters.

Images