A method and device for dispatching medium-voltage lines in a distribution network

Abstract

This invention relates to the field of power digitization technology and discloses a method and apparatus for dispatching medium-voltage lines in a distribution network. The method includes: constructing a topology map of the distribution network in the area to be dispatched; selecting lines to be predicted; obtaining the switching states of all switching devices within the line at K sampling times within the time period [t-K,t) to form a target state sequence set; obtaining a target average current value based on the average current value of all non-switching devices in the line to be predicted within the time period [t-K,t); searching for candidate historical time periods within the time period [1,t-K]; calculating the Pearson similarity between the historical state sequence set of each candidate historical time period and the target state sequence set; obtaining the switching states of all switching devices in the candidate historical time period corresponding to the one with the highest Pearson similarity; and dispatching all switching devices on the line to be predicted in the distribution network in the area to be dispatched at K sampling times after the current time.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power digitization, and in particular to a method and device for dispatching medium-voltage lines in a distribution network. Background Art

[0002] There are many factors that affect the dispatching and operation strategy of medium-voltage lines in the distribution network, including transmission and distribution. However, as the transmission capacity gradually stabilizes, the dispatching strategy is increasingly closely related to power demand. In order to improve the safe and stable operation of the power grid, methods such as orderly power consumption and advance quota have been implemented. However, with the rapid promotion and access of distributed new energy and the diversified growth of power demand, the power dispatching strategy is facing the test of new emergencies.

[0003] The grid line topology is the skeleton of the grid. The power dispatching strategy determines the update and change of the grid line topology. Therefore, the dispatching strategy under the current power consumption situation is deduced based on the sequence changes of the historical line topology to provide assistance for the decision-making of the dispatching strategy. However, the changes in the historical line topology are difficult to express. If it is directly described, there are technical difficulties in the topology comparison between multiple versions. In addition, data from different sources have differences in format, accuracy and timeliness, and need to be standardized for effective integration. However, the process of data standardization is often complicated and time-consuming.

[0004] At the same time, the scheduling strategy of a single switch / control type node in the topology is difficult to predict and has strong randomness. Moreover, due to the change in the state of a single switch / control type node, the topology structure will continue to change, making the formulation of scheduling strategies based on line topology extremely complicated. Traditional complex models based on artificial intelligence technologies such as neural networks that predict strategies have high computing resource requirements and are difficult to provide scheduling strategies in a timely and efficient manner. There is a lag in line scheduling, which leads to reduced grid stability. Summary of the invention

[0005] To this end, the technical problem to be solved by the present invention is to overcome the problem in the prior art that it is difficult to predict the state sequence of the line in a timely and accurate manner based on the complex and changeable distribution network topology structure that is directly described, and thus it is impossible to timely and accurately dispatch the distribution network.

[0006] In order to solve the above technical problems, the present invention provides a method for dispatching medium-voltage lines in a distribution network, comprising:

[0007] Taking the devices in the distribution network of the area to be dispatched as nodes, a topological diagram of the distribution network of the area to be dispatched is constructed; the devices include switch type devices and non-switch type devices;

[0008] Select the line to be predicted in the topological diagram of the distribution network in the area to be dispatched, and obtain the During the time period The switch states at the sampling moments constitute the target state sequence set; Indicates the current moment;

[0009] Based on the prediction of each non-switching device in the line During the time period The sampled current value at the sampling time is calculated for each non-switching device in the predicted line. The average current value in the time period is obtained to obtain the current of all non-switching devices in the line to be predicted. The average value of the average current values ​​within the time period is taken as the target average current value;

[0010] exist During the time period, find all non-switching devices in the predicted line. The average value of the average current value at the sampling time is equal to the target average current value, which is used as the candidate historical time period; each candidate historical time period includes sampling moments;

[0011] For each candidate historical time period, obtain The switch states of all switch-type devices at the sampling moments constitute a historical state sequence set; based on the Pearson similarity coefficient, the Pearson similarity between the historical state sequence set of each candidate historical time period and the target state sequence set is calculated;

[0012] Get the candidate historical time period corresponding to the maximum Pearson similarity , ;by During the time period The switch states of all switch-type devices in the line to be predicted at the sampling time are taken as a set of predicted state sequences;

[0013] Based on the set of predicted state sequences, During the time period All switch-type devices in the line to be predicted are scheduled at each sampling moment.

[0014] Preferably, if The candidate historical time periods cannot be obtained during the time period, including:

[0015] exist During the time period, find all non-switching devices in the predicted line. The average value of the average current value at the sampling time is equal to the target average current value, and is used as a candidate historical long time period; each candidate historical long time period includes sampling moments; among them, is a positive integer not less than 2;

[0016] Get each candidate historical long time period The switch states of all switch-type devices at a sampling moment constitute a long historical state sequence set;

[0017] The dynamic time warping algorithm is used to calculate the curve similarity between the long historical state sequence set of each candidate long historical time period and the target state sequence set, and the candidate long historical time period with the largest curve similarity is obtained. , ;

[0018] by During the time period The switch states of all switch-type devices in the line to be predicted at the sampling time are taken as a long prediction state sequence set;

[0019] Based on the long prediction state sequence set, the prediction of the line in the distribution network of the dispatching area is During the time period All switch type devices are scheduled at each sampling moment.

[0020] Preferably, the dynamic time warping algorithm is used to calculate the curve similarity between the long history state sequence set of each candidate long history time period and the target state sequence set, which is expressed as:

[0021] ;

[0022] in, Indicates A set of long history state sequences of candidate long historical time periods and The target state sequence set of time periods The similarity of the curves between , Indicates the total number of candidate long-term historical time periods; long-term historical state sequence set , Indicates that the line to be predicted is The first candidate historical long period The long history state sequence at time There are a total of candidate historical long periods sampling moments; target state sequence set , Indicates that the line to be predicted is In the time period The target state sequence at each moment; Represents the Dynamic Time Warping algorithm.

[0023] Preferably, a topological map of the distribution network in the area to be dispatched is constructed based on the archive data of the equipment in the distribution network in the area to be dispatched; the archive data of the equipment includes basic archive data and operation measurement data;

[0024] The basic file data includes equipment type, equipment subtype, voltage level, equipment line, equipment power station, terminal information, longitude and latitude coordinates, power supply area and commissioning date;

[0025] The operation measurement data includes equipment type, equipment subtype, voltage level, comprehensive multiple, measurement value type, measurement time and measurement value.

[0026] Preferably, the construction of a topological map of the distribution network in the area to be dispatched based on the archive data of the equipment in the distribution network in the area to be dispatched includes:

[0027] Based on the archive data of each device, obtain the terminal information of each device;

[0028] If the two terminal numbers stored in the terminal information of any two devices have the same terminal number, then the two devices are physically connected, and a connection edge is established between the two nodes representing the two devices;

[0029] The terminal information of all devices is traversed until connection edges are established between all nodes represented by two devices with the same terminal number, the topology map of the distribution network in the area to be dispatched is initialized, and the topology map of the distribution network in the area to be dispatched is obtained.

[0030] Preferably, the line to be predicted is selected in the topological diagram of the distribution network in the area to be scheduled, and the prediction of all switch-type devices in the line to be predicted is obtained. During the time period The switch states at the sampling moments constitute the target state sequence set, including:

[0031] Get all the switch type devices in the line to be predicted The switch state at each sampling moment in the time period constitutes the target state sequence corresponding to the sampling moment;

[0032] based on The target state sequences at all sampling moments within the time period constitute the target state sequence set;

[0033] The switch state of the switch type device is represented by a measurement value. When the measurement value is 1, the switch state of the switch type device is on, and when the measurement value is 0, the switch state of the switch type device is off.

[0034] Preferably, the method is based on each non-switching device in the line to be predicted. During the time period The sampled current value at the sampling time is calculated for each non-switching device in the predicted line. The average current value during a time period is expressed as:

[0035] ;

[0036] in, Indicates the first Non-switching devices exist The average current value during the time period, Indicates the first Non-switching equipment The sampling current value at the sampling moment; , express The total number of samples in the time period; Indicates the first A non-switching device, , Indicates the total number of non-switching devices in the line to be predicted.

[0037] Preferably, the target average current value is expressed as:

[0038] ;

[0039] in, Indicated in The target average current value of the line to be predicted within the time period.

[0040] Preferably, the Pearson similarity coefficient is used to calculate the Pearson similarity between the historical state sequence set of each candidate historical time period and the target state sequence set, which is expressed as:

[0041] ;

[0042] in, Indicates Candidate historical time periods The historical state sequence set and the target state sequence set The Pearson similarity between , Indicates the total number of candidate historical time periods; historical state sequence set , , and They represent the lines to be predicted in Candidate historical time periods The first sampling time, Sampling time and The historical state sequence at the sampling time, Candidate historical time periods shared within The sampling time, ; Indicates that the line to be predicted is The average value of historical states in candidate historical time periods is expressed as: ; Target state sequence set , Indicates that the line to be predicted is In the time period The target state sequence at each moment; Indicates that the line to be predicted is The average value of the target state within the time period, expressed as: .

[0043] This embodiment provides a medium-voltage line dispatching device for a distribution network, including:

[0044] A topology map construction module, using the devices in the distribution network of the area to be dispatched as nodes to construct a topology map of the distribution network of the area to be dispatched; the devices include switch type devices and non-switch type devices;

[0045] The target state sequence set acquisition module is used to select the line to be predicted in the topological diagram of the distribution network in the area to be dispatched, and obtain the state of all switch-type devices in the line to be predicted. During the time period The switch states at the sampling moments constitute the target state sequence set; Indicates the current moment;

[0046] The target average current value acquisition module is used to obtain the target average current value based on the average current of each non-switching device in the line to be predicted. During the time period The sampled current value at the sampling time is calculated for each non-switching device in the predicted line. The average current value in the time period is obtained to obtain the current of all non-switching devices in the line to be predicted. The average value of the average current values ​​within the time period is taken as the target average current value;

[0047] The candidate historical time period acquisition module is used to During the time period, find all non-switching devices in the predicted line. The average value of the average current value at the sampling time is equal to the target average current value, which is used as the candidate historical time period; each candidate historical time period includes sampling moments;

[0048] The prediction state sequence set acquisition module is used to obtain, for each candidate historical time period, The switch states of all switch-type devices at the sampling moment constitute a historical state sequence set; based on the Pearson similarity coefficient, the Pearson similarity between the historical state sequence set of each candidate historical time period and the target state sequence set is calculated; the candidate historical time period corresponding to the maximum Pearson similarity is obtained , ;by During the time period The switch states of all switch-type devices in the line to be predicted at the sampling time are taken as a set of predicted state sequences;

[0049] The scheduling module is used to During the time period All switch-type devices in the line to be predicted are scheduled at each sampling moment.

[0050] The above technical solution of the present invention has the following beneficial effects compared with the prior art:

[0051] The method for dispatching medium-voltage lines in the distribution network described in the present invention, after constructing the topological map of the distribution network in the area to be dispatched, performs sequence feature analysis on the topological map, and constructs a state sequence set using the state of the switch-type equipment in the selected line to be predicted within a time period; based on the state sequence change law reflected in the state sequence set of the medium-voltage line of the distribution network, the average current value of the line is used as a comparison basis to find a predicted state sequence set, and predict the dispatching strategy within the time period after the current moment; the present invention uses the state sequence composed of the switch state of the switch-type equipment in the line for prediction, which can directly reflect the essential law of switch control in the medium-voltage line of the distribution network, so that the predicted dispatching strategy is more timely and accurate, has stronger applicability, and improves the reliability and stability of power supply. At the same time, when the present invention uses the Pearson similarity coefficient to calculate the similarity, the demand for computing resources is not high, and lightweight deployment can be carried out to meet the fast computing requirements of the dispatching strategy and ensure the real-time dispatching.

[0052] When faced with state sequences of unequal lengths, the present invention uses the dynamic time warping algorithm DTW to calculate the similarity between two state sequence sets; the dynamic time warping algorithm DTW pays more attention to the changing trend of the sequence, and can also identify sequences with higher similarity in a set of state sequences with inconsistent lengths, thereby improving the matching accuracy, improving the accuracy of state sequence prediction, and optimizing the operating efficiency and stability of the power grid based on accurate scheduling strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below according to specific embodiments of the present invention in conjunction with the accompanying drawings, wherein:

[0054] Figure 1 It is a flow chart of the steps of a method for dispatching medium-voltage lines in a distribution network provided by the present invention;

[0055] Figure 2 It is a schematic diagram of initializing the topology map provided by the present invention;

[0056] Figure 3 It is a schematic diagram of the timing selection of the line topology scheduling control strategy provided by the present invention. DETAILED DESCRIPTION

[0057] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0058] Reference Figure 1 As shown, a flow chart of the steps of a method for dispatching medium-voltage lines in a distribution network provided by the present invention includes the following specific steps:

[0059] S101: constructing a topological map of the distribution network in the area to be dispatched by using devices in the area to be dispatched as nodes; the devices include switch type devices and non-switch type devices;

[0060] S102: Select a line to be predicted in the topological diagram of the distribution network in the area to be dispatched, and obtain the information of all switch-type devices in the line to be predicted. During the time period The switch states at the sampling moments constitute the target state sequence set; Indicates the current moment;

[0061] S103: Based on the prediction of each non-switch type device in the line During the time period The sampled current value at the sampling time is calculated for each non-switching device in the predicted line. The average current value in the time period is obtained to obtain the current of all non-switching devices in the line to be predicted. The average value of the average current values ​​within the time period is taken as the target average current value;

[0062] S104: During the time period, find all non-switching devices in the predicted line. The average value of the average current value at the sampling time is equal to the target average current value, which is used as the candidate historical time period; each candidate historical time period includes sampling moments;

[0063] S105: For each candidate historical time period, obtain The switch states of all switch-type devices at the sampling moments constitute a historical state sequence set; based on the Pearson similarity coefficient, the Pearson similarity between the historical state sequence set of each candidate historical time period and the target state sequence set is calculated;

[0064] S106: Obtain the candidate historical time period corresponding to the maximum Pearson similarity , ;by During the time period The switch states of all switch-type devices in the line to be predicted at the sampling time are taken as a set of predicted state sequences;

[0065] S107: Based on the predicted state sequence set, During the time period All switch-type devices in the line to be predicted are scheduled at each sampling moment.

[0066] Specifically, in step S104, if The candidate historical time periods cannot be obtained during the time period, including:

[0067] S108: During the time period, find all non-switching devices in the predicted line. The average value of the average current value at the sampling time is equal to the target average current value, and is used as a candidate historical long time period; each candidate historical long time period includes sampling moments; among them, is a positive integer not less than 2;

[0068] S109: Obtain each candidate historical long time period The switch states of all switch-type devices at a sampling moment constitute a long historical state sequence set;

[0069] S110: Calculate the curve similarity between the long historical state sequence set of each candidate long historical time period and the target state sequence set using a dynamic time warping algorithm, and obtain the candidate long historical time period with the largest curve similarity , ;

[0070] S111: During the time period The switch states of all switch-type devices in the line to be predicted at the sampling time are taken as a long prediction state sequence set;

[0071] S112: Based on the long prediction state sequence set, the predicted line in the distribution network of the dispatching area is During the time period All switch type devices are scheduled at each sampling moment.

[0072] Among them, A set of long history state sequences of candidate long historical time periods and The target state sequence set of time periods The curve similarity between is expressed as:

[0073] ;

[0074] , Indicates the total number of candidate long-term historical time periods; long-term historical state sequence set , Indicates that the line to be predicted is The first candidate historical long period The long history state sequence at time There are a total of candidate historical long periods sampling moments; target state sequence set , Indicates that the line to be predicted is In the time period The target state sequence at each moment; Represents the Dynamic Time Warping algorithm.

[0075] When faced with state sequences of unequal lengths, the present invention uses the dynamic time warping algorithm DTW to calculate the similarity between two state sequence sets; the dynamic time warping algorithm DTW pays more attention to the changing trend of the sequence, and can also identify sequences with higher similarity in a set of state sequences with inconsistent lengths, thereby improving the matching accuracy, improving the accuracy of state sequence prediction, and optimizing the operating efficiency and stability of the power grid based on accurate scheduling strategies.

[0076] In an embodiment of the present invention, a topological diagram of the distribution network in the area to be dispatched is constructed based on the archive data of the equipment in the distribution network in the area to be dispatched; the archive data of the equipment includes basic archive data and operation measurement data. The basic archive data includes equipment type, equipment subtype, voltage level, equipment-to-which line, equipment-to-which power station, terminal information, longitude and latitude coordinates, power supply area and commissioning date; the operation measurement data includes equipment type, equipment subtype, voltage level, comprehensive multiple, measurement value type, measurement time and measurement value.

[0077] Specifically, the non-switch type equipment includes distribution transformers and special transformers, and the switch type equipment includes fuses, disconnectors and load switches.

[0078] Specifically, refer to Figure 2 As shown, it is a schematic diagram of topology initialization; in step S101, based on the archive data of the equipment in the regional distribution network to be dispatched, a topology map of the regional distribution network to be dispatched is constructed, including:

[0079] S101-1: Based on the archive data of each device, obtain the terminal information of each device;

[0080] S101-2: If the two terminal numbers stored in the terminal information of any two devices have the same terminal number, the two devices are physically connected, and a connection edge is established between two nodes representing the two devices;

[0081] S101-3: Traverse the terminal information of all devices until connection edges are established between all nodes represented by two devices with the same terminal number, complete the initialization of the topological map of the distribution network in the area to be dispatched, and obtain the topological map of the distribution network in the area to be dispatched.

[0082] Specifically, in step S102, the acquisition of the target state sequence set includes:

[0083] S102-1: Obtain all switch type devices in the line to be predicted The switch state at each sampling moment in the time period constitutes the target state sequence corresponding to the sampling moment;

[0084] S102-2: Based on The target state sequences at all sampling moments within the time period constitute the target state sequence set;

[0085] The switch state of the switch type device is represented by a measurement value. When the measurement value is 1, the switch state of the switch type device is on. When the measurement value is 0, the switch state of the switch type device is off.

[0086] Specifically, in step S103, obtaining the target average current value includes:

[0087] S103-1: The method based on each non-switch type device in the line to be predicted During the time period The sampled current value at the sampling time is calculated for each non-switching device in the predicted line. The average current value during a time period is expressed as:

[0088] ;

[0089] in, Indicates the first Non-switching devices exist The average current value during the time period, Indicates the first Non-switching equipment The sampling current value at the sampling moment; , express The total number of samples in the time period; Indicates the first A non-switching device, , Indicates the total number of non-switching devices in the line to be predicted.

[0090] S103-2: Obtain all non-switch type devices in the line to be predicted The average value of the average current value within the time period is used as the target average current value and is expressed as:

[0091] ;

[0092] in, Indicated in The target average current value of the line to be predicted within the time period.

[0093] In the embodiment of the present invention, Candidate historical time periods The historical state sequence set and the target state sequence set The Pearson similarity between , expressed as:

[0094] ;

[0095] , Indicates the total number of candidate historical time periods; historical state sequence set , , and They represent the lines to be predicted in Candidate historical time periods The first sampling time, Sampling time and The historical state sequence at the sampling time, Candidate historical time periods shared within The sampling time, ; Indicates that the line to be predicted is The average value of historical states in candidate historical time periods is expressed as: ; Target state sequence set , Indicates that the line to be predicted is In the time period The target state sequence at each moment; Indicates that the line to be predicted is The average value of the target state within the time period, expressed as: .

[0096] Specifically, in step S107, During the time period When all the switch-type devices in the predicted line are scheduled at the sampling time, the predicted state sequence set During the time period The switch status of all switch-type devices in the line to be predicted at the sampling time is taken as the reference, corresponding to During the time period The switch states of all switch-type devices in the line to be predicted at the sampling time are set to During the time period The switch states of all switch-type devices in the line to be predicted are consistent at the sampling time, completing the During the time period The scheduling of all switch-type devices in the line to be predicted at a sampling time.

[0097] The method for dispatching medium-voltage lines in the distribution network described in the present invention, after constructing the topological map of the distribution network in the area to be dispatched, performs sequence feature analysis on the topological map, and constructs a state sequence set using the state of the switch-type equipment in the selected line to be predicted within a time period; based on the state sequence change law reflected in the state sequence set of the medium-voltage line of the distribution network, the average current value of the line is used as a comparison basis to find a predicted state sequence set, and predict the dispatching strategy within the time period after the current moment; the present invention uses the state sequence composed of the switch state of the switch-type equipment in the line for prediction, which can directly reflect the essential law of switch control in the medium-voltage line of the distribution network, so that the predicted dispatching strategy is more timely and accurate, has stronger applicability, and improves the reliability and stability of power supply. At the same time, when the present invention uses the Pearson similarity coefficient to calculate the similarity, the demand for computing resources is not high, and lightweight deployment can be carried out to meet the fast computing requirements of the dispatching strategy and ensure the real-time dispatching.

[0098] Based on the above embodiments, in an embodiment of the present invention, the medium-voltage line dispatching method of the distribution network provided by the embodiment of the present invention is used to dispatch resources for the target area power grid, specifically including:

[0099] S201: Select a target area power grid and obtain production equipment archive data in the target area distribution network;

[0100] Production equipment archive data, including: basic archive data and measurement data of transformers and switches connected to 10KV lines in the target area;

[0101] Refer to Table 1, which is the basic file data of the equipment, including equipment number, equipment type, equipment subtype, voltage level, affiliated line, affiliated power station, terminal number, longitude and latitude coordinates, power supply area, and commissioning date;

[0102] Table 1 Basic file data of equipment

[0103]

[0104] Refer to Table 2, which shows the measurement data of the equipment; wherein, the measurement data of the switch equipment records the operating state of the switch at each measurement time, 1 represents that the switch equipment is in a closed state, and 0 represents that the switch equipment is in an open state; the measurement data of the switch equipment records the electrical operating parameters of the equipment, including voltage, current, power and other data, and each measurement value represents the electrical parameter value of the equipment measured at that moment. Among them, the measurement data of non-switch equipment includes voltage, current, power and other data; since the line voltage is in a stable state, the influence on the topology change law is small, therefore, in this embodiment, the current measurement value is selected to analyze the overall topology change law of the line.

[0105] Table 2 Measurement data of equipment

[0106]

[0107] S202: Use a topology graph model to describe the topology structure of the medium voltage line, and construct a node for each device in the topology graph.

[0108] The topology diagram is initialized according to the terminal information of the device. If the terminal information of device N1 is [7025SR09222, 7025SR09226], and the terminal information of device N2 is [7025SR33425, 7025SR09226], at this time, the two have the same terminal number 7025SR09226, so it is considered that the two are physically connected.

[0109] Accordingly, in the topology graph, connection edges are established for the nodes corresponding to the two devices; successively, all devices in the topology graph are connected to complete the initialization of the topology graph;

[0110] S203: Selecting a time range and line topology for policy recommendation;

[0111] Reference Figure 3 As shown, it is a schematic diagram of the timing selection of the line topology scheduling control strategy;

[0112] For example, the prediction target is [ t 0, t 1] The state sequence of the topological structure of a certain line in the period S 0,1 , select the time t 0 to time t Topological sequence in the period of 2 P 2,0 =[ p 1, p 2,..., p k ],time t 0 and time t 2 are standard collection time points. The average current value of all distribution transformers in the target line during this period is calculated and expressed as:

[0113] ;

[0114] in, Indicates the number of current acquisitions during this period; Indicates transformer During the target period [ t 2, t 0] within the average current value; Indicates transformer The current value at time i; the topology is [ t 2, t 0] the average current value during the period, expressed as: ;in, Indicates the total number of all distribution transformers in the target line topology;

[0115] S204: Obtain the measurement data of the topology segment, and start from the earliest time t a Start, select a time period of the same length to the end time t b ( t b - t a = t 0- t 2) Calculate the average current value of the topology during this period;

[0116] If the average current value during this period With target period [ t 2, t 0], that is, , record the topological sequence within the period, and define the topological sequence corresponding to the c periods [ P 1,1 , P 1,2 ,..., P 1,c ] meets the above requirements, P 1,1 = Pa , b =[ p 1,1,a , p 1,1,a+1 ,..., p 1,1,b ];

[0117] If no equal time period is found after the search is completed, the time period T is set to 2T, and the search is repeated until a time period with an average current value equal to the average current value of the target time period is found, and the topological state sequence of all topological nodes in the searched time period is recorded; the measurement data of all switches in it are taken out, and a total of d time periods are found [ P 2,1 , P 2,2 ,..., P 2,c ], meeting the above requirements;

[0118] S205: If an equal time period is found within the T time period, the Pearson similarity coefficient is used to calculate the similarity between the two and find the one with the greatest similarity;

[0119] Pearson similarity, expressed as:

[0120] ;

[0121] in, and Respectively represent [ t 0, t 1] Time period and [ t 2, t 0] The average current value during the period is expressed as , ;

[0122] Obtain the topological state sequence within the time period corresponding to the maximum value of the similarity coefficient and output it as the predicted state sequence;

[0123] S206: If the time period is 2T, the measurement sequence is calculated using the dynamic time warping algorithm P 2 and P 2,0 The curve similarity of is expressed as: ;

[0124] Obtain the topological state sequence within the time period corresponding to the maximum value of the similarity coefficient and output it as the predicted state sequence;

[0125] S207: Obtain the output prediction state sequence and dispatch the target area power grid;

[0126] The present invention aims at the bottleneck problem of high complexity and difficulty in automatic generation in the formulation of medium-voltage topology dispatching and operation strategies in distribution networks, and introduces the topology graph sequence feature analysis technology to solve and implement it. The present invention uses the historical topology state change sequence of the medium-voltage line as a data sample to extract the topological dynamic change law of the line. The similarity between the historical topological sequence and the current topological sequence is calculated based on the current average value, so as to predict the future dispatching strategy of the switch in the target line. The graph model is used to describe the dynamic change sequence of the topological state, and the calculation efficiency is high. Since the control strategy of the medium-voltage line switch is most affected by the power load and the physical characteristics of the line itself, the present invention predicts the dispatching strategy by extracting the line topological state change law. Compared with the traditional method of predicting the line dispatching control strategy based on the power load, it can more directly reflect the essential law of switch control in the line topology and has better applicability. In addition, the formulation of the dispatching control strategy has a high real-time requirement. Compared with the complex model that uses artificial intelligence technologies such as neural networks to predict the strategy, the present method does not require high computing resources, and can carry out lightweight deployment and implementation applications to meet the needs of fast dispatching calculations.

[0127] Based on the above embodiments, an embodiment of the present invention further provides a medium-voltage line dispatching device for a distribution network, comprising:

[0128] A topology map construction module 100 constructs a topology map of the distribution network in the area to be dispatched, using the devices in the distribution network in the area to be dispatched as nodes; the devices include switch type devices and non-switch type devices;

[0129] The target state sequence set acquisition module 200 is used to select the line to be predicted in the topological diagram of the distribution network in the area to be scheduled, and obtain the state of all switch-type devices in the line to be predicted. During the time period The switch states at the sampling moments constitute the target state sequence set; Indicates the current moment;

[0130] The target average current value acquisition module 300 is used to obtain the target average current value based on the average current value of each non-switching device in the line to be predicted. During the time period The sampled current value at the sampling time is calculated for each non-switching device in the predicted line. The average current value in the time period is obtained to obtain the current of all non-switching devices in the line to be predicted. The average value of the average current values ​​within the time period is taken as the target average current value;

[0131] The candidate historical time period acquisition module 400 is used to During the time period, find all non-switching devices in the predicted line. The average value of the average current value at the sampling time is equal to the target average current value, which is used as the candidate historical time period; each candidate historical time period includes sampling moments;

[0132] The prediction state sequence set acquisition module 500 is used to obtain, for each candidate historical time period, The switch states of all switch-type devices at the sampling moment constitute a historical state sequence set; based on the Pearson similarity coefficient, the Pearson similarity between the historical state sequence set of each candidate historical time period and the target state sequence set is calculated; the candidate historical time period corresponding to the maximum Pearson similarity is obtained , ;by During the time period The switch states of all switch-type devices in the line to be predicted at the sampling time are taken as a set of predicted state sequences;

[0133] The scheduling module 600 is used to schedule During the time period All switch-type devices in the line to be predicted are scheduled at each sampling moment.

[0134] The medium-voltage line dispatching device of the distribution network of the present embodiment is used to implement the aforementioned medium-voltage line dispatching method of the distribution network. Therefore, the specific implementation method of the medium-voltage line dispatching device of the distribution network can be seen in the implementation method part of the medium-voltage line dispatching method of the distribution network in the previous text. For example, the topology map construction module 100, the target state sequence set acquisition module 200, the target average current value acquisition module 300, and the candidate historical time period acquisition module 400 are respectively used to implement steps S101, S102, S103 and S104 in the above-mentioned medium-voltage line dispatching method of the distribution network; the predicted state sequence set acquisition module 500 is used to implement steps S105 and S106 in the above-mentioned medium-voltage line dispatching method of the distribution network; the dispatching module 600 is used to implement step S107 in the above-mentioned medium-voltage line dispatching method of the distribution network; therefore, its specific implementation method can refer to the description of the corresponding embodiments of each part, which will not be repeated here.

[0135] The method for dispatching medium-voltage lines in the distribution network described in the present invention, after constructing the topological map of the distribution network in the area to be dispatched, performs sequence feature analysis on the topological map, and constructs a state sequence set using the state of the switch-type equipment in the selected line to be predicted within a time period; based on the state sequence change law reflected in the state sequence set of the medium-voltage line of the distribution network, the average current value of the line is used as a comparison basis to find a predicted state sequence set, and predict the dispatching strategy within the time period after the current moment; the present invention uses the state sequence composed of the switch state of the switch-type equipment in the line for prediction, which can directly reflect the essential law of switch control in the medium-voltage line of the distribution network, so that the predicted dispatching strategy is more timely and accurate, has stronger applicability, and improves the reliability and stability of power supply. At the same time, when the present invention uses the Pearson similarity coefficient to calculate the similarity, the demand for computing resources is not high, and lightweight deployment can be carried out to meet the fast computing requirements of the dispatching strategy and ensure the real-time dispatching. When faced with state sequences of unequal lengths, the present invention uses the dynamic time warping algorithm DTW to calculate the similarity between two state sequence sets; the dynamic time warping algorithm DTW pays more attention to the changing trend of the sequence, and can also identify sequences with higher similarity in a set of state sequences with inconsistent lengths, thereby improving the matching accuracy, improving the accuracy of state sequence prediction, and optimizing the operating efficiency and stability of the power grid based on accurate scheduling strategies.

[0136] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0137] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0138] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0139] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0140] Obviously, the above embodiments are merely examples for the purpose of clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the present invention.

Claims

1. A method for dispatching medium voltage lines in a distribution network, characterized in that: include: Taking the equipment in the distribution network of the area to be dispatched as nodes, a topological map of the distribution network of the area to be dispatched is constructed; The equipment includes switch type equipment and non-switch type equipment; Select the line to be predicted in the topological diagram of the distribution network in the area to be dispatched, and obtain the During the time period The switch states at the sampling moments constitute the target state sequence set; Indicates the current moment; Based on the prediction of each non-switching device in the line During the time period The sampled current value at the sampling time is calculated for each non-switching device in the predicted line. The average current value in the time period is obtained to obtain the current of all non-switching devices in the line to be predicted. The average value of the average current values ​​within the time period is taken as the target average current value; exist During the time period, find all non-switching devices in the predicted line. The average value of the average current value at the sampling time is equal to the target average current value, which is used as the candidate historical time period; each candidate historical time period includes sampling moments; For each candidate historical time period, obtain The switch states of all switch-type devices at the sampling moments constitute a historical state sequence set; based on the Pearson similarity coefficient, the Pearson similarity between the historical state sequence set of each candidate historical time period and the target state sequence set is calculated; Get the candidate historical time period corresponding to the maximum Pearson similarity , ;by During the time period The switch states of all switch-type devices in the line to be predicted at the sampling time are taken as a set of predicted state sequences; Based on the set of predicted state sequences, During the time period All switch-type devices in the line to be predicted are scheduled at each sampling moment.

2. The method for dispatching medium voltage lines in a distribution network according to claim 1, characterized in that: If in The candidate historical time periods cannot be obtained during the time period, including: exist During the time period, find all non-switching devices in the predicted line. The average value of the average current value at the sampling time is equal to the target average current value, and is used as a candidate historical long time period; each candidate historical long time period includes sampling moments; among them, is a positive integer not less than 2; Get each candidate historical long time period The switch states of all switch-type devices at a sampling moment constitute a long historical state sequence set; The dynamic time warping algorithm is used to calculate the curve similarity between the long historical state sequence set of each candidate long historical time period and the target state sequence set, and the candidate long historical time period with the largest curve similarity is obtained. , ; by During the time period The switch states of all switch-type devices in the line to be predicted at the sampling time are taken as a long prediction state sequence set; Based on the long prediction state sequence set, the prediction of the line in the distribution network of the dispatching area is During the time period All switch type devices are scheduled at each sampling moment.

3. The method for dispatching medium voltage lines in a distribution network according to claim 2, characterized in that: The dynamic time warping algorithm is used to calculate the curve similarity between the long history state sequence set of each candidate long history period and the target state sequence set, which is expressed as: ； in, Indicates A set of long history state sequences of candidate long historical time periods and The target state sequence set of time periods The similarity of the curves between , Indicates the total number of candidate long-term historical time periods; long-term historical state sequence set , Indicates that the line to be predicted is The first candidate historical long period The long history state sequence at time There are a total of candidate historical long periods sampling moments; target state sequence set , Indicates that the line to be predicted is In the time period The target state sequence at each moment; Represents the Dynamic Time Warping algorithm.

4. The method for dispatching medium voltage lines in a distribution network according to claim 1, characterized in that: Based on the archive data of the equipment in the distribution network of the area to be dispatched, a topological map of the distribution network in the area to be dispatched is constructed; the archive data of the equipment includes basic archive data and operation measurement data; The basic file data includes equipment type, equipment subtype, voltage level, equipment line, equipment power station, terminal information, longitude and latitude coordinates, power supply area and commissioning date; The operation measurement data includes equipment type, equipment subtype, voltage level, comprehensive multiple, measurement value type, measurement time and measurement value.

5. The method for dispatching medium voltage lines in a distribution network according to claim 4, characterized in that: The method of constructing a topological map of the distribution network in the area to be dispatched based on the archive data of the equipment in the distribution network in the area to be dispatched includes: Based on the archive data of each device, obtain the terminal information of each device; If the two terminal numbers stored in the terminal information of any two devices have the same terminal number, then the two devices are physically connected, and a connection edge is established between the two nodes representing the two devices; The terminal information of all devices is traversed until connection edges are established between all nodes represented by two devices with the same terminal number, the topology map of the distribution network in the area to be dispatched is initialized, and the topology map of the distribution network in the area to be dispatched is obtained.

6. The method for dispatching medium voltage lines in a distribution network according to claim 4, characterized in that: The method selects a line to be predicted in the topological diagram of the distribution network in the area to be scheduled, obtains the current status of all switch-type devices in the line to be predicted, and During the time period The switch states at the sampling moments constitute the target state sequence set, including: Get all the switch type devices in the line to be predicted The switch state at each sampling moment in the time period constitutes the target state sequence corresponding to the sampling moment; based on The target state sequences at all sampling moments within the time period constitute the target state sequence set; The switch state of the switch type device is represented by a measurement value. When the measurement value is 1, the switch state of the switch type device is on, and when the measurement value is 0, the switch state of the switch type device is off.

7. The method for dispatching medium voltage lines in a distribution network according to claim 1, characterized in that: The method is based on each non-switching device in the line to be predicted. During the time period The sampled current value at the sampling time is calculated for each non-switching device in the predicted line. The average current value during a time period is expressed as: ； in, Indicates the first Non-switching devices exist The average current value during the time period, Indicates the first Non-switching equipment The sampling current value at the sampling moment; , express The total number of samples in the time period; Indicates the first A non-switching device, , Indicates the total number of non-switching devices in the line to be predicted.

8. The method for dispatching medium voltage lines in a distribution network according to claim 7, characterized in that: The target average current value is expressed as: ； in, Indicated in The target average current value of the line to be predicted within the time period.

9. The method for dispatching medium voltage lines in a distribution network according to claim 1, characterized in that: Based on the Pearson similarity coefficient, the Pearson similarity between the historical state sequence set of each candidate historical time period and the target state sequence set is calculated, which is expressed as: ； in, Indicates Candidate historical time periods The historical state sequence set and the target state sequence set The Pearson similarity between , Indicates the total number of candidate historical time periods; historical state sequence set , , and They represent the lines to be predicted in Candidate historical time periods The first sampling time, Sampling time and The historical state sequence at the sampling time, Candidate historical time periods shared within The sampling time, ; Indicates that the line to be predicted is The average value of historical states in candidate historical time periods is expressed as: ; Target state sequence set , Indicates that the line to be predicted is In the time period The target state sequence at each moment; Indicates that the line to be predicted is The average value of the target state within the time period, expressed as: .

10. A medium voltage line dispatching device for a distribution network, characterized in that: include: A topology map construction module uses the devices in the distribution network of the area to be dispatched as nodes to construct a topology map of the distribution network of the area to be dispatched; The equipment includes switch type equipment and non-switch type equipment; The target state sequence set acquisition module is used to select the line to be predicted in the topological diagram of the distribution network in the area to be dispatched, and obtain the state of all switch-type devices in the line to be predicted. During the time period The switch states at the sampling moments constitute the target state sequence set; Indicates the current moment; The target average current value acquisition module is used to obtain the target average current value based on the average current of each non-switching device in the line to be predicted. During the time period The sampled current value at the sampling time is calculated for each non-switching device in the predicted line. The average current value in the time period is obtained to obtain the current of all non-switching devices in the line to be predicted. The average value of the average current values ​​within the time period is taken as the target average current value; The candidate historical time period acquisition module is used to During the time period, find all non-switching devices in the predicted line. The average value of the average current value at the sampling time is equal to the target average current value, which is used as the candidate historical time period; each candidate historical time period includes sampling moments; The prediction state sequence set acquisition module is used to obtain, for each candidate historical time period, The switch states of all switch-type devices at the sampling moments constitute a historical state sequence set; based on the Pearson similarity coefficient, the Pearson similarity between the historical state sequence set of each candidate historical time period and the target state sequence set is calculated; Get the candidate historical time period corresponding to the maximum Pearson similarity , ;by During the time period The switch states of all switch-type devices in the line to be predicted at the sampling time are taken as a set of predicted state sequences; The scheduling module is used to During the time period All switch-type devices in the line to be predicted are scheduled at each sampling moment.

Images