A voltage regulation power supply control method and system for a multifunctional power supply

Abstract

The present invention relates to a voltage regulation power supply control method and system for a multifunctional power supply, and relates to the field of substation power distribution. The method comprises: when the power grid is not out of power, obtaining real-time power grid status information, combining the power distribution equipment position number and the demand control mode, statistically analyzing the voltage fluctuation trend of the same-mode power grid, obtaining a first voltage regulation control parameter, and starting the power grid power access circuit to access the power grid and execute the power distribution request based on the first voltage regulation control parameter; when the power grid is out of power, obtaining the scheduled load status of the UPS power supply, statistically analyzing the voltage fluctuation trend of the same-mode UPS power supply, performing voltage regulation control optimization, obtaining a second voltage regulation control parameter, and starting the UPS power access circuit to access the UPS power supply and execute the power distribution request. The method solves the technical problem in the prior art that, during the construction of the substation AC / DC panel renovation project, the DC power supply system needs to be moved, which easily causes the associated protection equipment to lose power, thereby affecting the operation of the substation.

Description

Technical Field

[0001] The present invention relates to the field of power distribution in transformer substations, and in particular to a voltage regulation and power supply control method and system for a multifunctional power supply. Background Art

[0002] During the construction of the substation AC / DC panel renovation project, the DC power supply system needs to be moved. However, since the DC panel uses large and heavy equipment, once a problem occurs with the DC equipment, all related protection devices will lose power, which is difficult to restore in a short time and will directly affect the safe operation of the substation.

[0003] The current solution is to monitor the DC power system in real time during the construction of the substation AC / DC panel renovation project and to promptly perform maintenance when an abnormality occurs. However, the disadvantage is that the DC power system has poor controllability when moved, and there is still an abnormality risk. Summary of the Invention

[0004] The present invention aims to solve the technical problem in the prior art that during the construction of the AC / DC panel renovation project of a substation, the DC power supply system needs to be moved, which easily causes the related protection equipment to lose power, thereby affecting the operation of the substation. A voltage regulation power supply control method and system for a multifunctional power supply are provided to solve the problem.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] In a first aspect, the present invention provides a voltage regulation power supply control method for a multifunctional power supply, which is applied to a multifunctional power supply voltage regulation power supply device, wherein the multifunctional power supply voltage regulation power supply device includes a grid power supply access circuit and a UPS power supply access circuit, including:

[0007] Receive a power distribution request uploaded by a user, wherein the power distribution request includes a power distribution equipment position number and a demand control mode;

[0008] When the power grid is not disconnected, real-time grid status information is obtained, and based on the power distribution equipment position number and the demand control mode, the voltage fluctuation trend of the same-mode grid is statistically analyzed, and voltage regulation control optimization is performed to obtain a first voltage regulation control parameter, and the grid power supply access circuit is started to access the grid and execute the power distribution request based on the first voltage regulation control parameter;

[0009] When the power grid is out of power, the scheduled load status of the UPS power supply is obtained, and the voltage fluctuation trend of the UPS power supply with the same mode is statistically analyzed in combination with the position number of the distribution equipment and the demand control mode. The voltage regulation control is optimized to obtain the second voltage regulation control parameter, and the UPS power supply access circuit is started to access the UPS power supply to execute the power distribution request based on the second voltage regulation control parameter.

[0010] Furthermore, when the power grid is not cut off, the real-time state information of the power grid is obtained, and the voltage fluctuation trend of the same-mode power grid is statistically calculated in combination with the position number of the power distribution equipment and the demand control mode, including:

[0011] Obtaining a load-side circuit topology including the position number of the power distribution equipment;

[0012] The load-side circuit topology and the real-time state information of the power grid are used as load pre-connection constraint conditions, and the power distribution equipment position number and the demand control mode are used as load connection constraint conditions;

[0013] Retrieve a grid distribution sample set that satisfies the load pre-connection constraint condition and the load connection constraint condition through the Internet;

[0014] The centroid of the voltage fluctuation trend of the power distribution grid sample set is extracted to obtain the voltage fluctuation trend of the same-mode power grid.

[0015] Furthermore, searching the network for a power grid distribution sample set that satisfies the load pre-connection constraint condition and the load connection constraint condition includes:

[0016] Retrieving a first power grid distribution sample set that satisfies the load pre-connection constraint condition and the load connection constraint condition, wherein any power grid distribution sample in the first power grid distribution sample set includes first power grid real-time state information and a first demand control mode;

[0017] updating the load pre-connection constraint condition based on the first power grid real-time state information, updating the load access constraint condition based on the first demand control mode, and retrieving a second power grid distribution sample set that satisfies the updated load pre-connection constraint condition and the updated load access constraint condition;

[0018] The first power grid power distribution sample set and the second power grid power distribution sample set are added to the power grid power distribution sample set.

[0019] Furthermore, the load-side circuit topology and the real-time state information of the power grid are used as load pre-connection constraint conditions, and the power distribution equipment position number and the demand control mode are used as load connection constraint conditions, including:

[0020] Construct a load-side circuit topology deviation evaluation function:

[0021] ,

[0022] in, Characterizes the deviation value between the sample load-side circuit topology and the load-side circuit topology, The i-th node representing the same topological position of the sample load-side circuit topology and the load-side circuit topology, Characterizes the degree difference between the sample load-side circuit topology and the i-th node of the load-side circuit topology, Represents the branch formed by node x and node y, Characterize the branch set, Characterize the same branch in the topological position of the sample load-side circuit topology and the load-side circuit topology The impedance difference, Characterization weight coefficient, determined based on the Delphi method;

[0023] Construct a function to evaluate the deviation of real-time state information of power grid:

[0024] ,

[0025] in, Characterizes the deviation evaluation value between the real-time state information of the sample power grid and the real-time state information of the power grid, Characterize the k-th attribute grid real-time state parameter, Characterize the normalized deviation of the k-th attribute grid real-time state parameter between the sample grid real-time state information and the grid real-time state information, Characterizes the normalized deviation tolerance value of the predefined k-th attribute grid real-time state parameter, Characterize the preset weight of the k-th attribute power grid real-time state parameter, The total number of parameter attributes representing the real-time status;

[0026] Construct a function to evaluate the deviation of the power distribution equipment tag number:

[0027] ,

[0028] ,

[0029] in, Characterizes the deviation evaluation value between the sample distribution equipment number and the distribution equipment number, The model representing the tag number of the sample power distribution equipment, The model that represents the position number of the power distribution equipment, Characterize the sample distribution equipment number, Characterizes the position number of the power distribution equipment, Characterizes the normalized value of the service time deviation between the sample distribution equipment number and the distribution equipment number, Characterizes the error tolerance value of the normalized value of service time deviation, Characterize the topological distribution location of the sample power distribution equipment number, Characterize the topological distribution location of the power distribution equipment number, Characterizes the Euclidean distance between the sample distribution equipment position number and the topological distribution position of the distribution equipment position number, Characterizes the Euclidean distance deviation threshold, 、 and Characterization weight coefficient, determined based on the Delphi method;

[0030] Construct the demand control modal deviation evaluation function:

[0031] ,

[0032] in, Characterizes the deviation evaluation value between the sample demand control mode and the demand control mode, Characterizes the control mode phase angle deviation between the sample demand control mode and the demand control mode, Characterizes the control parameter vector deviation, Characterization weight coefficient, determined based on the Delphi method;

[0033] Constructing the load pre-connection constraint condition according to the load-side circuit topology deviation evaluation function and the first deviation threshold, and the grid real-time state information deviation evaluation function and the second deviation threshold;

[0034] The load access constraint condition is constructed based on the power distribution equipment bit number deviation evaluation function and the third deviation threshold, and the demand control modal deviation evaluation function and the fourth deviation threshold.

[0035] Furthermore, the voltage fluctuation trend of the power distribution grid sample set is subjected to centroid extraction to obtain the voltage fluctuation trend of the same-mode power grid, including:

[0036] Clustering the second power grid distribution sample set according to the first power grid distribution sample set to obtain multiple clusters of power grid distribution samples;

[0037] Traversing the plurality of clusters of power distribution grid samples to extract the LOF outlier factor minimum voltage fluctuation trend, and setting it as a plurality of first centroid voltage fluctuation trends;

[0038] extracting a plurality of second voltage fluctuation trends from the first power grid distribution sample set;

[0039] LOF outlier factor minimum voltage fluctuation trend extraction is performed on the multiple first centroid voltage fluctuation trends and the multiple second voltage fluctuation trends to obtain the same-mode power grid voltage fluctuation trend.

[0040] Furthermore, the voltage fluctuation trend of the same-mode power grid is statistically analyzed, and voltage regulation control optimization is performed to obtain the first voltage regulation control parameter, including:

[0041] When the voltage variance of the voltage fluctuation trend of the same-mode power grid is greater than or equal to a voltage variance threshold, performing uniform distribution based on the rated interval of the voltage regulation control parameter to obtain a plurality of voltage regulation control parameters;

[0042] By using a voltage regulation simulator bound to a grid access node, the plurality of voltage regulation control parameters are traversed and processed in combination with the same-mode grid voltage fluctuation trend to obtain a plurality of grid voltage fluctuation correction trends;

[0043] When extracting a grid voltage fluctuation correction trend whose voltage variance is less than the voltage variance threshold value from the plurality of grid voltage fluctuation correction trends, the trend is set as the first voltage regulation control parameter.

[0044] Furthermore, it also includes:

[0045] When the number of grid voltage fluctuation correction trends whose voltage variance is less than the voltage variance threshold is equal to 0, sorting the voltage regulation control parameters in ascending order of voltage variance to obtain a voltage regulation control parameter sequence;

[0046] Extracting the first three voltage regulation control parameters of the voltage regulation control parameter sequence and performing same-attribute mean configuration to obtain target voltage regulation control parameters;

[0047] Taking the target voltage regulation control parameter as the regulation target, performing Euclidean distance reduction adjustment on the last 50% of the voltage regulation control parameters in the voltage regulation control parameter sequence to obtain updated voltage regulation control parameters;

[0048] When the voltage variance of the updated voltage regulation control parameter is less than the grid voltage fluctuation correction trend of the voltage variance threshold, the first voltage regulation control parameter is set; otherwise, the iteration is continued.

[0049] Furthermore, a voltage regulation simulator bound to the grid access node is used to traverse the voltage regulation control parameters and process them in combination with the same-mode grid voltage fluctuation trend to obtain several grid voltage fluctuation correction trends, including:

[0050] Collecting voltage regulation control logs of the grid access node based on the voltage regulator model, wherein the voltage regulation control logs include initial grid voltage fluctuation trend record data, voltage regulation control parameter record data, and post-regulation grid voltage fluctuation trend record data;

[0051] Configuring a voltage regulation simulator architecture, wherein the voltage regulation simulator architecture includes a first input branch based on a bidirectional gated recurrent network, a second input branch based on a fully connected layer, a backbone network composed of a feature concatenation layer and a temporal convolutional network connected in sequence, and an output layer based on a time-distributed fully connected layer;

[0052] The voltage regulation simulator is trained using the recorded data of the voltage fluctuation trend of the adjusted power grid as supervision and the recorded data of the initial power grid voltage fluctuation trend and the recorded data of the voltage regulation control parameters as input.

[0053] Furthermore, when extracting the grid voltage fluctuation correction trend whose voltage variance is less than the voltage variance threshold from the plurality of grid voltage fluctuation correction trends, setting it as the first voltage regulation control parameter includes:

[0054] When the voltage variance of the updated voltage regulation control parameter is less than the grid voltage fluctuation correction trend of the voltage variance threshold, adding the voltage regulation control parameter to be selected;

[0055] The candidate voltage regulation control parameters are sorted based on minimum energy consumption to obtain the first voltage regulation control parameter.

[0056] In a second aspect, the present invention provides a voltage regulation power supply control system for a multifunctional power supply, which is applied to a multifunctional power supply voltage regulation power supply device, wherein the multifunctional power supply voltage regulation power supply device includes a grid power supply access circuit and a UPS power supply access circuit, including:

[0057] A power distribution request receiving module, configured to receive a power distribution request uploaded by a user, wherein the power distribution request includes a power distribution equipment position number and a demand control mode;

[0058] A first power distribution control module is configured to, when the power grid is not disconnected, obtain real-time power grid status information, combine the power distribution equipment position number and the demand control mode, calculate the voltage fluctuation trend of the same-mode power grid, perform voltage regulation control optimization, obtain a first voltage regulation control parameter, start the power grid power supply access circuit to connect to the power grid, and execute the power distribution request based on the first voltage regulation control parameter;

[0059] The second power distribution control module is used to obtain the scheduled load status of the UPS power supply when the power grid is out of power, combine the power distribution equipment position number and the demand control mode, count the voltage fluctuation trend of the UPS power supply with the same mode, perform voltage regulation control optimization, obtain the second voltage regulation control parameter, start the UPS power supply access circuit to access the UPS power supply and execute the power distribution request based on the second voltage regulation control parameter.

[0060] The beneficial effect of the present invention is that the operation of the DC power supply system during construction is replaced by configuring a multifunctional power supply voltage regulating power supply device, the multifunctional power supply voltage regulating power supply device includes a grid power access circuit and a UPS power access circuit, when the grid is not powered off, the grid power access circuit is started to access the grid, when the grid is powered off, the UPS power access circuit is started to access the UPS power supply, in order to ensure the stability of the supply distribution voltage, according to the real-time status information of the grid, combined with the position number of the distribution equipment and the demand control mode, the same-mode grid voltage fluctuation trend and the same-mode UPS power supply voltage fluctuation trend are analyzed to pre-configure the voltage regulation parameters, and then the distribution control is performed according to the pre-set voltage regulation parameters, thereby achieving the technical effect of ensuring stable power distribution during the construction of the substation DC panel. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A schematic flow chart of a voltage regulation and power supply control method for a multifunctional power supply provided by the present invention;

[0062] Figure 2 A schematic diagram of a flow chart of determining a homomodal grid voltage fluctuation trend in a voltage regulation power supply control method for a multifunctional power supply provided by the present invention. DETAILED DESCRIPTION

[0063] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0064] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the specified features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0065] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or illustration". Any embodiment of the present invention described as "for example" is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed herein.

[0066] Example 1, as Figure 1 As shown, an embodiment of the present invention provides a voltage regulation power supply control method for a multifunctional power supply, which is applied to a multifunctional power supply voltage regulation power supply device, wherein the multifunctional power supply voltage regulation power supply device includes a grid power supply access circuit and a UPS power supply access circuit, including the following steps:

[0067] S10: receiving a power distribution request uploaded by a user, wherein the power distribution request includes a power distribution equipment position number and a demand control mode;

[0068] Specifically, a multifunctional power supply voltage regulation and power supply control method provided in an embodiment of the present application needs to be implemented through a multifunctional power supply voltage regulation and power supply device. The multifunctional power supply voltage regulation and power supply device includes a grid power supply access circuit and a UPS power supply access circuit. The grid power supply access circuit is an access line connected to the mains power, which realizes AC-DC conversion through an inverter, thereby performing DC 0V-220V debugging work during the construction process; the UPS power supply access circuit refers to a line connected to the UPS power supply, which outputs a DC power supply of 0V-220V through rectification and inversion. Based on the redundancy principle, the two power supplies are isolated and operated, and the normal operation of the transmission equipment can still be guaranteed when one power supply fails. Preferably, the multifunctional power supply voltage regulation and power supply device integrates a group of lithium-ion battery packs. When the external power supply is input, the device charges the built-in lithium-ion battery pack (charged together when the quick interface is connected to the external battery pack), and supplies power to the transmission equipment at the same time.

[0069] During the specific power supply, when the user needs to perform voltage regulation test on the equipment during the construction process and has power distribution demand, a power distribution request will be sent to the multi-functional power supply voltage regulation power supply device through the user end, such as the mobile phone portable terminal and the central control screen. Optionally, a pre-set distribution equipment position number is selected through the menu bar. The position number is bound to the fixed distribution equipment model and the service life of the distribution equipment. The demand control mode refers to the specific parameters that need to be controlled, exemplarily the output voltage, output frequency and other parameters of the distribution equipment position number.

[0070] Through the power distribution request uploaded by the user end, the software control module of the multi-functional power supply voltage regulation and power supply device can parse the power distribution equipment position number and demand control mode, determine the required power distribution data, and facilitate subsequent refined control of power distribution.

[0071] S20: When the power grid is not disconnected, obtain real-time status information of the power grid, combine the position number of the power distribution equipment and the demand control mode, calculate the voltage fluctuation trend of the power grid with the same mode, perform voltage regulation control optimization, obtain a first voltage regulation control parameter, start the power grid power supply access circuit to connect to the power grid, and execute the power distribution request based on the first voltage regulation control parameter;

[0072] Specifically, when the power grid is not cut off, that is, the mains power can be used for voltage regulation test distribution work. At this time, there is no need to start the UPS power supply. However, since the stability of the power supply process will also affect the accuracy of the voltage regulation test results, it is necessary to statistically analyze the voltage fluctuation trend of the power grid under the same control state according to the power distribution equipment position number and the demand control mode, and obtain the voltage fluctuation trend of the same-mode power grid; further, according to the predicted same-mode power grid voltage fluctuation trend, the voltage compensation control parameters are optimized to obtain the pre-configured voltage regulation control parameters, and then voltage regulation and distribution are carried out. This can ensure the voltage stability of the subsequent voltage regulation test process and avoid the influence of voltage fluctuations on the test results.

[0073] Preferably, the voltage compensation control parameters include proportional gain: directly responding to the instantaneous value of the voltage deviation and quickly generating a control signal proportional to the deviation. When the voltage deviates from the target value (such as a voltage sag), the proportional gain immediately increases the compensation current or reactive power output to shorten the adjustment time; integral gain: accumulating historical deviations to eliminate steady-state errors (such as continuous voltage offsets caused by long-term load fluctuations); differential gain: reducing overshoot and oscillation during the adjustment process through negative feedback of the voltage change rate; filter cutoff frequency: the frequency threshold of the low-pass filter, and signal components above this frequency are attenuated; response time: the time required from detecting the voltage deviation to the output compensation amount reaching 90% of the target value.

[0074] In detail, the above steps include two key contents. The first is: according to the real-time status information of the power grid, combined with the position number of the distribution equipment and the demand control mode, the voltage fluctuation trend of the same-mode power grid is counted; the second is to count the voltage fluctuation trend of the same-mode power grid, perform voltage regulation control optimization, and obtain the first voltage regulation control parameter.

[0075] First, regarding the first aspect: based on the real-time status information of the power grid, combined with the position number of the power distribution equipment and the demand control mode, the voltage fluctuation trend of the same-mode power grid is statistically analyzed. The implementation algorithm provided in the embodiment of the present application is as follows:

[0076] Further, such as Figure 2As shown, when the power grid is not cut off, the real-time state information of the power grid is obtained, and the voltage fluctuation trend of the same-mode power grid is counted in combination with the position number of the distribution equipment and the demand control mode. Step S20 includes the following steps:

[0077] S21: Obtain a load-side circuit topology including the position number of the power distribution equipment;

[0078] S22: Using the load-side circuit topology and the real-time state information of the power grid as load pre-connection constraint conditions, and using the power distribution equipment position number and the demand control mode as load connection constraint conditions;

[0079] S23: searching online for a power grid distribution sample set that meets the load pre-connection constraint condition and the load connection constraint condition;

[0080] S24: performing centroid extraction on the voltage fluctuation trend of the power distribution grid sample set to obtain the voltage fluctuation trend of the same-mode power grid.

[0081] Specifically, the load-side circuit topology refers to the circuit structure on the load side to which the distribution equipment position number that needs to be connected to the power grid belongs; the real-time status information of the power grid refers to the real-time status of the power grid before connecting to the power grid, preferably including the voltage timing information, current timing information, active power, reactive power, power factor, etc. of the power grid, which can be customized by the management end. After the setting is completed, the corresponding sensor is deployed, and the sensor is connected to the voltage regulating power supply device for communication to realize full monitoring of the real-time status information of the power grid.

[0082] When the power grid is not cut off, the load-side circuit topology and the real-time status information of the power grid are used as the constraints before the load is connected, and the distribution equipment position number and the demand control mode are used as the constraints for the load connection. That is, the load-side circuit topology connected to the power grid is the same, and the real-time status information of the power grid before connection is the same, and the distribution equipment position number of the load side connected to the power grid is the same, and the demand control mode of the distribution equipment position number is the same. The historical distribution samples are stored as a power grid distribution sample set. Each power grid distribution sample includes the voltage fluctuation trend of a preset time after the power supply is connected to the power grid. The preset time is the time required for power supply preset by the user.

[0083] Preferably, the load side circuit topology connected to the power grid is the same, the real-time state information of the power grid before access is the same, the power distribution equipment position number of the load side connected to the power grid is the same, and the demand control mode of the power distribution equipment position number is the same. The sorting process of the historical power distribution samples is as follows:

[0084] Furthermore, with the load-side circuit topology and the real-time state information of the power grid as load pre-connection constraint conditions, and with the power distribution equipment position number and the demand control mode as load connection constraint conditions, step S222 includes the following steps:

[0085] S2221: Construct a load-side circuit topology deviation evaluation function:

[0086] ,

[0087] in, Characterizes the deviation value between the sample load-side circuit topology and the load-side circuit topology, The i-th node representing the same topological position of the sample load-side circuit topology and the load-side circuit topology, Characterizes the degree difference between the sample load-side circuit topology and the i-th node of the load-side circuit topology, Represents the branch formed by node x and node y, Characterize the branch set, Characterize the same branch in the topological position of the sample load-side circuit topology and the load-side circuit topology The impedance difference, Characterization weight coefficient, determined based on the Delphi method;

[0088] S2222: Construct a function to evaluate the deviation of real-time state information of power grid:

[0089] ,

[0090] in, Characterizes the deviation evaluation value between the real-time state information of the sample power grid and the real-time state information of the power grid, Characterize the k-th attribute grid real-time state parameter, Characterize the normalized deviation of the k-th attribute grid real-time state parameter between the sample grid real-time state information and the grid real-time state information, Characterizes the normalized deviation tolerance value of the predefined k-th attribute grid real-time state parameter, Characterize the preset weight of the k-th attribute power grid real-time state parameter, The total number of parameter attributes representing the real-time status;

[0091] S2223: Construct a function to evaluate the deviation of the power distribution equipment bit number:

[0092] ,

[0093] ,

[0094] in, Characterizes the deviation evaluation value between the sample distribution equipment number and the distribution equipment number, The model representing the tag number of the sample power distribution equipment, The model that represents the position number of the power distribution equipment, Characterize the sample distribution equipment number, Characterizes the position number of the power distribution equipment, Characterizes the normalized value of the service time deviation between the sample distribution equipment number and the distribution equipment number, Characterizes the error tolerance value of the normalized value of service time deviation, Characterize the topological distribution location of the sample power distribution equipment number, Characterize the topological distribution location of the power distribution equipment number, Characterizes the Euclidean distance between the sample distribution equipment position number and the topological distribution position of the distribution equipment position number, Characterizes the Euclidean distance deviation threshold, 、 and Characterization weight coefficient, determined based on the Delphi method;

[0095] S2224: Construct demand control modal deviation evaluation function:

[0096] ,

[0097] in, Characterizes the deviation evaluation value between the sample demand control mode and the demand control mode, Characterizes the control mode phase angle deviation between the sample demand control mode and the demand control mode, Characterizes the control parameter vector deviation, Characterization weight coefficient, determined based on the Delphi method;

[0098] S2225: Constructing the load pre-connection constraint condition according to the load-side circuit topology deviation evaluation function and the first deviation threshold, and the grid real-time state information deviation evaluation function and the second deviation threshold;

[0099] S2226: Construct the load access constraint condition according to the power distribution equipment position number deviation evaluation function and the third deviation threshold, and the demand control mode deviation evaluation function and the fourth deviation threshold.

[0100] Specifically, a load-side circuit topology deviation evaluation function is constructed: , used to realize the topological structure deviation of the sample and load-side circuit topology; construct the grid real-time state information deviation evaluation function: , used to assess deviations in real-time grid status information. Pre-load access constraints are considered satisfied when the deviation assessment value between the sample load-side circuit topology and the load-side circuit topology is less than or equal to the user-set first deviation threshold (default is 2.5, depending on the user's tolerance for topology deviation), and the deviation assessment value between the sample real-time grid status information and the real-time grid status information is less than or equal to the user-set second deviation threshold (default is 1.2, depending on the user's tolerance for real-time grid status deviation).

[0101] Furthermore, a function for evaluating the position deviation of power distribution equipment is constructed: , , used to evaluate the deviation between the sample distribution equipment bit number and the distribution equipment bit number, and to construct the demand control modal deviation evaluation function: , used to evaluate the deviation between the sample demand control mode and the demand control mode. When the deviation between the sample distribution equipment bit number and the distribution equipment bit number is less than or equal to the user-defined third deviation threshold (the default is 0.8, depending on the user's tolerance for distribution equipment deviation), and the deviation between the sample demand control mode and the demand control mode is less than or equal to the user-defined fourth deviation threshold (the default is 0.5, depending on the user's tolerance for demand control mode deviation, especially: model mismatch is directly rejected, and the output is greater than 0.5), the load access constraint is considered to be met.

[0102] Furthermore, the collection process of the power grid distribution sample set is as follows:

[0103] Furthermore, the network is searched for a power grid distribution sample set that meets the load pre-connection constraint condition and the load connection constraint condition. Step S22 includes the following steps:

[0104] S221: Retrieve a first power grid distribution sample set that meets the load pre-connection constraint condition and the load connection constraint condition, wherein any power grid distribution sample in the first power grid distribution sample set includes first power grid real-time state information and a first demand control mode;

[0105] S222: updating the load pre-connection constraint based on the first power grid real-time status information, updating the load access constraint based on the first demand control mode, and retrieving a second power grid distribution sample set that satisfies the updated load pre-connection constraint and the updated load access constraint.

[0106] S223: Add the first power grid power distribution sample set and the second power grid power distribution sample set into the power grid power distribution sample set.

[0107] Specifically, first, a first power grid distribution sample set that satisfies both the load pre-access constraint and the load access constraint is retrieved from local storage data or networked power grid data; further, to avoid insufficient sample data, which may lead to a lack of accuracy in the analysis results, the embodiment of the present application proposes a strategy of adjacency analysis, and uses the first power grid distribution sample set as a constraint basis to collect samples to obtain a second power grid distribution sample set, thereby expanding the samples to solve the problem of insufficient samples, and at the same time ensuring that the deviation from the load-side circuit topology, the real-time status information of the power grid, the position number of the distribution equipment, and the demand control mode is small. Preferably, the process of determining the second power grid distribution sample set includes:

[0108] Extract the first power grid real-time status information and the first demand control mode of any power grid distribution sample, that is, the parameters used by the first power grid distribution sample to participate in the constraint condition comparison, replace the power grid real-time status information and the demand control mode of the load access constraint condition and the load pre-access constraint condition with the first power grid real-time status information and the first demand control mode, obtain the updated load pre-access constraint condition and the updated load access constraint condition, then retrieve the second power grid distribution sample set that meets the updated load pre-access constraint condition and the updated load access constraint condition, add the first power grid distribution sample set and the second power grid distribution sample set to the power grid distribution sample set, store them in a pending response state, and wait for the next step to quickly call them.

[0109] Preferably, the aforementioned sample collection process has been described. Now, the centroid extraction process is described as follows:

[0110] Furthermore, the centroid of the voltage fluctuation trend of the power distribution grid sample set is extracted to obtain the voltage fluctuation trend of the same-mode power grid. Step S24 includes the following steps:

[0111] S241: Clustering the second power grid distribution sample set according to the first power grid distribution sample set to obtain multiple clusters of power grid distribution samples;

[0112] S242: Traversing the plurality of clusters of power distribution grid samples to extract LOF outlier factor minimum voltage fluctuation trends, and setting them as a plurality of first centroid voltage fluctuation trends;

[0113] S243: Extracting multiple second voltage fluctuation trends from the first power grid distribution sample set;

[0114] S244: Perform LOF outlier factor minimum voltage fluctuation trend extraction on the multiple first centroid voltage fluctuation trends and the multiple second voltage fluctuation trends to obtain the same-mode power grid voltage fluctuation trend.

[0115] Specifically, first, since the second power grid distribution sample set is collected separately for each sample in the first power grid distribution sample set, the second power grid distribution sample set can be clustered according to the first power grid distribution sample set to obtain multiple clusters of power grid distribution samples; further, the multiple clusters of power grid distribution samples are traversed to extract the voltage fluctuation trend of the minimum LOF outlier factor of each cluster, and set it as multiple first centroid voltage fluctuation trends;

[0116] Then, multiple second voltage fluctuation trends of the first power grid distribution sample set are extracted; further, the multiple first centroid voltage fluctuation trends and the multiple second voltage fluctuation trends are subjected to LOF outlier factor minimum voltage fluctuation trend extraction to obtain the same-mode power grid voltage fluctuation trend.

[0117] In detail, the LOF outlier factor calculation process for each voltage fluctuation trend is as follows, using an arbitrary cluster of power distribution samples as an example:

[0118] As can be seen from the above, the time step of each voltage fluctuation trend is the same, so the voltage at each moment of each voltage fluctuation trend can be used as the coordinate of a dimension to obtain the coordinate of each voltage fluctuation trend; then the coordinate Euclidean distance of the pairwise voltage fluctuation trends of the clusters to be analyzed is calculated; further, 1 / 3 of the total number of clusters is rounded up as the neighborhood number threshold, and for any voltage fluctuation trend in the cluster, a number of voltage fluctuation trends equal to the neighborhood number threshold are selected from near to far based on the coordinate Euclidean distance, and the inverse of the mean of the Euclidean distance between the several voltage fluctuation trends and the corresponding voltage fluctuation trend is calculated, and stored as the local density of the corresponding voltage fluctuation trend; the local density is calculated for each voltage fluctuation trend in the cluster; the mean of all local densities in the cluster is calculated, and then the LOF outlier factor value of each voltage fluctuation trend is obtained by comparing the mean with the local density of each voltage fluctuation trend, so that the voltage fluctuation trend with the minimum LOF outlier factor can be screened out. The smaller the outlier factor, the more likely it is the centroid within the cluster, so it can be extracted as the output result.

[0119] Secondly, for the second aspect: statistically analyzing the voltage fluctuation trend of the same-mode power grid, performing voltage regulation control optimization, and obtaining the first voltage regulation control parameter, the implementation algorithm provided in the embodiment of the present application is as follows:

[0120] Furthermore, the voltage fluctuation trend of the same-mode power grid is statistically analyzed, and voltage regulation control optimization is performed to obtain the first voltage regulation control parameter. Step S20 includes the following steps:

[0121] S25: When the voltage variance of the voltage fluctuation trend of the same-mode power grid is greater than or equal to the voltage variance threshold, performing uniform distribution based on the rated interval of the voltage regulation control parameter to obtain a plurality of voltage regulation control parameters;

[0122] S26: traversing the plurality of voltage regulation control parameters through a voltage regulation simulator bound to the grid access node, processing the plurality of voltage regulation control parameters in combination with the same-mode grid voltage fluctuation trends, and obtaining a plurality of grid voltage fluctuation correction trends;

[0123] S27: Extracting a grid voltage fluctuation correction trend whose voltage variance is less than the voltage variance threshold from the plurality of grid voltage fluctuation correction trends and setting it as the first voltage regulation control parameter.

[0124] Specifically, as previously mentioned, voltage compensation control parameters include proportional gain, integral gain, differential gain, filter cutoff frequency, and response time. The rated range of voltage regulation control parameters refers to the specific rated range of the voltage compensation control parameters. This rated range is limited by the device performing voltage regulation, and different devices have a factory-set rated range for voltage regulation control parameters. Several voltage regulation control parameters refer to the results of uniformly distributed value selection within the rated range of the voltage regulation control parameters. The detailed implementation is as follows:

[0125] Assuming that the voltage compensation control parameters include proportional gain, integral gain, differential gain, filter cutoff frequency, and response time, a high-dimensional space with five dimensions including proportional gain, integral gain, differential gain, filter cutoff frequency, and response time can be constructed; then any position in the high-dimensional space is equivalent to a voltage regulation control parameter. The user can pre-set the number of uniform distributions, with the default value being 500. Based on the number of 500, the initial solution is distributed in the high-dimensional space using a uniform distribution function and stored as several voltage regulation control parameters. Preferably, the convergence rule for determining whether the distribution is uniform is set as follows:

[0126] Calculate the ratio of the distribution interval length of each dimensional parameter of the 500 randomly generated solutions to the rated interval length, and calculate the mean of multiple dimensional ratios and store them as the distribution occupancy rate. Then calculate the variance of the distribution distance between every two solutions. If the variance of the distribution distance is less than or equal to the user-preset variance threshold, and the distribution occupancy rate is greater than or equal to the user-preset distribution occupancy rate threshold, it is considered that the uniform distribution has converged, and several voltage regulation control parameters are output. Otherwise, continue to update and perform cyclic judgment.

[0127] The voltage regulation simulator is a functional component used to process voltage regulation control parameters and the voltage fluctuation trend of the same-mode grid, and to predict the voltage fluctuation trend of the grid after voltage regulation. In particular, the voltage regulation simulator is uniquely bound to the grid access node. The detailed construction process is as follows:

[0128] Furthermore, by traversing the plurality of voltage regulation control parameters through a voltage regulation simulator bound to the grid access node, and processing them in combination with the same-mode grid voltage fluctuation trend, a plurality of grid voltage fluctuation correction trends are obtained. Step S26 includes the following steps:

[0129] S261: Collecting a voltage regulation control log of the grid access node based on a voltage regulator model, wherein the voltage regulation control log includes initial grid voltage fluctuation trend record data, voltage regulation control parameter record data, and post-regulation grid voltage fluctuation trend record data;

[0130] S262: Configuring a voltage regulation simulator architecture, wherein the voltage regulation simulator architecture includes a first input branch based on a bidirectional gated recurrent network, a second input branch based on a fully connected layer, a backbone network, and an output layer based on a time-distributed fully connected layer, wherein the backbone network is composed of a feature concatenation layer and a temporal convolutional network connected in sequence;

[0131] S263: Using the recorded data of the voltage fluctuation trend of the adjusted power grid as supervision, and using the recorded data of the initial power grid voltage fluctuation trend and the recorded data of the voltage adjustment control parameters as input, the voltage adjustment simulator is trained.

[0132] Specifically, the network architecture for configuring the voltage regulation simulator is as follows:

[0133] [voltage timing input] → first input branch → eigenvector;

[0134] [Control parameter input] → second input branch → condition vector;

[0135] [Feature vector, conditional vector] → backbone network [feature concatenation layer (time step replication and concatenation to ensure the consistency of the dimensions of the feature vector and conditional vector) → temporal convolutional network];

[0136] Backbone network [feature concatenation layer → temporal convolutional network] → temporal distributed fully connected layer → predicted voltage output.

[0137] Since the embodiment of the present application generally predicts short-term time series data, a lightweight time convolutional network is adopted to reduce computational complexity. Preferably, the first input branch is a bidirectional gated recurrent network, including 64 hidden units, 32 in each direction, and the second input branch is a fully connected layer, including 64 neurons. The time convolutional network includes 64 convolution kernels, and the output layer is configured as a time-distributed fully connected layer, including 1 neuron.

[0138] Preferably, the loss function is: ,in, Characterize the voltage regulation simulation loss value, Characterize the recorded data of the voltage fluctuation trend of the power grid after regulation, Characterize the grid voltage fluctuation trend prediction data, Characterize the weighted mean square error, Characterize the dynamic time warping distance, Characterize the weight parameters, , n represents the preset number of training times of fitting error, the default is 50 times, The weight representing the zth sample is a learning parameter, and the initial sample weights are the same; , A represents all regular paths that meet endpoint alignment and time monotonicity, v represents any pair of regular paths that meet endpoint alignment and time monotonicity, k represents the k-th moment characteristic value of a certain grid voltage fluctuation trend of any pair of regular paths that meet endpoint alignment and time monotonicity, l represents the l-th moment characteristic value of another grid voltage fluctuation trend of any pair of regular paths that meet endpoint alignment and time monotonicity, through Constrain the voltage value error at each time step to ensure absolute accuracy, through The predicted waveform is allowed to shift / stretch slightly on the time axis, but the overall fluctuation pattern is forced to be consistent, thereby ensuring the stability of the output results. As shown in the following table, the voltage regulation control log of the grid access node is collected with the voltage regulator model as the constraint. The voltage regulation control log includes the initial grid voltage fluctuation trend record data, the voltage regulation control parameter record data, and the grid voltage fluctuation trend record data after regulation:

[0139] Table 1: Training data example table

[0140]

[0141] Kp: proportional gain; Ki: integral gain; Kd: differential gain; fc (Hz): filter cutoff frequency; Tr (ms): response time.

[0142] The voltage regulation simulator is trained using the recorded data of the voltage fluctuation trend of the adjusted grid as supervision and the recorded data of the initial grid voltage fluctuation trend and the recorded data of the voltage regulation control parameters as input. Specifically, at least 50,000 pieces of training data are used, divided into training / validation / test sets according to a ratio of 70% / 15% / 15%. The voltage value is normalized to the nominal value of 220V, and the control parameters are normalized to the maximum allowable value. Then, training is started, and the convergence conditions are set as follows:

[0143] Table 2: Loss function convergence rules

[0144]

[0145] Preferably, the model is determined to have converged when the total loss fluctuation range of 10 consecutive samples is less than 1%.

[0146] Table 3: Model performance comparison

[0147]

[0148] Using the same loss function and sample data training, a single LSTM encoder + fully connected output LSTM-Only architecture, and CNN feature extraction + Transformer time series modeling CNN-Transformer, as well as a pure TCN architecture, the results show that the prediction model of the embodiment of the present application has lower loss, lower inference latency, and lower overshoot rate. Therefore, the model of the embodiment of the present application has higher accuracy. Based on each grid node that needs it, with the voltage regulator model as a constraint, the voltage regulation control log belonging to the grid access node is collected to ensure that the trained model has the ability to adapt to the scenario.

[0149] Furthermore, through the voltage regulation simulator bound to the grid access node, each voltage regulation control parameter of several voltage regulation control parameters and the same-mode grid voltage fluctuation trend are processed respectively, and several grid voltage fluctuation correction trends are output. Then, the variance value of the voltage values ​​at multiple moments of each grid voltage fluctuation correction trend is calculated. When the grid voltage fluctuation correction trend with a variance less than the voltage variance threshold set in advance by the user is extracted, it is set as the first voltage regulation control parameter for subsequent power supply control.

[0150] Furthermore, the method further includes step S28, which includes the following steps:

[0151] S281: When the number of grid voltage fluctuation correction trends whose voltage variance is less than the voltage variance threshold is equal to 0, sorting the voltage regulation control parameters in ascending order of voltage variance to obtain a voltage regulation control parameter sequence;

[0152] S282: extracting the first three voltage regulation control parameters of the voltage regulation control parameter sequence, performing same-attribute mean configuration, and obtaining a target voltage regulation control parameter;

[0153] S283: Taking the target voltage regulation control parameter as the regulation target, perform Euclidean distance reduction adjustment on the last 50% of the voltage regulation control parameters in the voltage regulation control parameter sequence to obtain updated voltage regulation control parameters;

[0154] S284: When the voltage variance of the updated voltage regulation control parameter is less than the grid voltage fluctuation correction trend of the voltage variance threshold, set it as the first voltage regulation control parameter; otherwise, continue iteration.

[0155] Specifically, the optimization process of voltage regulation control parameters is as follows:

[0156] When the number of grid voltage fluctuation correction trends whose voltage variance is less than the voltage variance threshold among the plurality of grid voltage fluctuation correction trends is equal to 0, the plurality of voltage regulation control parameters are sorted from small to large according to the voltage variance to obtain a voltage regulation control parameter sequence; the first three voltage regulation control parameters of the voltage regulation control parameter sequence are extracted and configured with the same attribute mean to obtain a target voltage regulation control parameter; with the target voltage regulation control parameter as the adjustment target, the last 50% of the voltage regulation control parameters in the voltage regulation control parameter sequence are subjected to Euclidean distance reduction adjustment to obtain an updated voltage regulation control parameter; when the voltage variance of the updated voltage regulation control parameter is less than the grid voltage fluctuation correction trend of the voltage variance threshold, it is set as the first voltage regulation control parameter; otherwise, the iteration is continued.

[0157] Euclidean distance reduction refers to the reduction in the voltage parameter distance between the regulated parameter and the regulation target at a certain moment. Using the top three voltage regulation control parameters as the parameters with better performance for the same attribute mean configuration, the target voltage regulation control parameters obtained can reduce contingency and ensure the global representativeness of the regulation target. The Euclidean distance reduction adjustment is performed on the last 50% of the voltage regulation control parameters to increase the probability and speed of obtaining the global optimal solution.

[0158] Furthermore, when extracting the grid voltage fluctuation correction trend whose voltage variance is less than the voltage variance threshold value from the plurality of grid voltage fluctuation correction trends, setting it as the first voltage regulation control parameter, step S27 includes the steps of:

[0159] S271: When the voltage variance of the updated voltage regulation control parameter is less than the grid voltage fluctuation correction trend of the voltage variance threshold, adding the voltage regulation control parameter to be selected;

[0160] S272: Perform minimum energy consumption sorting on the candidate voltage regulation control parameters to obtain the first voltage regulation control parameter.

[0161] Furthermore, when the voltage variance of the updated voltage regulation control parameter is less than the grid voltage fluctuation correction trend of the voltage variance threshold, the candidate voltage regulation control parameter is added, and the candidate voltage regulation control parameter is sorted for the minimum energy consumption to obtain the first voltage regulation control parameter. Generally speaking, the energy consumption value can be determined based on the size of the voltage regulation control parameter. Selecting the voltage regulation control parameter with lower energy consumption can ensure the energy-saving effect of power distribution.

[0162] S30: When the power grid is out of power, the scheduled load status of the UPS power supply is obtained, and the voltage fluctuation trend of the UPS power supply with the same mode is counted in combination with the position number of the distribution equipment and the demand control mode, and the voltage regulation control is optimized to obtain the second voltage regulation control parameter, and the UPS power supply access circuit is started to access the UPS power supply to execute the power distribution request based on the second voltage regulation control parameter.

[0163] Specifically, when the power grid is out of power, the scheduled load status of the UPS power supply is obtained, and the voltage fluctuation trend of the same-mode UPS power supply is statistically calculated in combination with the position number of the distribution equipment and the demand control mode; then the voltage regulation control optimization is implemented using the voltage regulation simulator bound to the UPS power supply model to obtain the second voltage regulation control parameter. Preferably, the same-mode UPS power supply voltage fluctuation trend and the construction process and implementation process of the voltage regulation simulator in the embodiment of the present application are exactly the same as the above, and no further details are given here. After obtaining the second voltage regulation control parameter, the UPS power supply access circuit is started to access the UPS power supply based on the second voltage regulation control parameter to execute the power distribution request to ensure power supply stability.

[0164] The embodiment of the present invention provides a method for controlling voltage regulation and power supply of a multifunctional power supply, which has at least the following technical effects:

[0165] By configuring a multifunctional power supply voltage regulating power supply device to replace the work of the DC power supply system during construction, the multifunctional power supply voltage regulating power supply device includes a grid power access circuit and a UPS power access circuit. When the grid is not powered off, the grid power access circuit is started to connect to the grid. When the grid is powered off, the UPS power access circuit is started to connect to the UPS power supply. In order to ensure the stability of the supply distribution voltage, according to the real-time status information of the grid, combined with the position number of the distribution equipment and the demand control mode, the same-mode grid voltage fluctuation trend and the same-mode UPS power supply voltage fluctuation trend are analyzed to pre-configure the voltage regulation parameters, and then the distribution control is performed according to the pre-set voltage regulation parameters, thereby achieving the technical effect of ensuring stable power distribution during the construction of the substation DC panel.

[0166] In a second embodiment, based on the same inventive concept as the voltage regulation power supply control method for a multifunctional power supply provided in the first embodiment, the present embodiment provides a voltage regulation power supply control system for a multifunctional power supply, which is applied to a multifunctional power supply voltage regulation power supply device. The multifunctional power supply voltage regulation power supply device includes a grid power supply access circuit and a UPS power supply access circuit, including:

[0167] A power distribution request receiving module, configured to receive a power distribution request uploaded by a user, wherein the power distribution request includes a power distribution equipment position number and a demand control mode;

[0168] A first power distribution control module is configured to, when the power grid is not disconnected, obtain real-time power grid status information, combine the power distribution equipment position number and the demand control mode, calculate the voltage fluctuation trend of the same-mode power grid, perform voltage regulation control optimization, obtain a first voltage regulation control parameter, start the power grid power supply access circuit to connect to the power grid, and execute the power distribution request based on the first voltage regulation control parameter;

[0169] The second power distribution control module is used to obtain the scheduled load status of the UPS power supply when the power grid is out of power, combine the power distribution equipment position number and the demand control mode, count the voltage fluctuation trend of the UPS power supply with the same mode, perform voltage regulation control optimization, obtain the second voltage regulation control parameter, start the UPS power supply access circuit to access the UPS power supply and execute the power distribution request based on the second voltage regulation control parameter.

[0170] Furthermore, when the power grid is not out of power, the real-time state information of the power grid is obtained, and the voltage fluctuation trend of the same-mode power grid is calculated by combining the position number of the power distribution equipment and the demand control mode. The execution steps include:

[0171] Obtaining a load-side circuit topology including the position number of the power distribution equipment;

[0172] The load-side circuit topology and the real-time state information of the power grid are used as load pre-connection constraint conditions, and the power distribution equipment position number and the demand control mode are used as load connection constraint conditions;

[0173] Retrieve a grid distribution sample set that satisfies the load pre-connection constraint condition and the load connection constraint condition through the Internet;

[0174] The centroid of the voltage fluctuation trend of the power distribution grid sample set is extracted to obtain the voltage fluctuation trend of the same-mode power grid.

[0175] Furthermore, the network is retrieved for a power grid distribution sample set that meets the load pre-connection constraint condition and the load connection constraint condition, and the execution steps include:

[0176] Retrieving a first power grid distribution sample set that satisfies the load pre-connection constraint condition and the load connection constraint condition, wherein any power grid distribution sample in the first power grid distribution sample set includes first power grid real-time state information and a first demand control mode;

[0177] updating the load pre-connection constraint condition based on the first power grid real-time state information, updating the load access constraint condition based on the first demand control mode, and retrieving a second power grid distribution sample set that satisfies the updated load pre-connection constraint condition and the updated load access constraint condition;

[0178] The first power grid power distribution sample set and the second power grid power distribution sample set are added to the power grid power distribution sample set.

[0179] Furthermore, taking the load-side circuit topology and the real-time state information of the power grid as load pre-connection constraint conditions, and taking the power distribution equipment position number and the demand control mode as load connection constraint conditions, the execution steps include:

[0180] Construct a load-side circuit topology deviation evaluation function:

[0181] ,

[0182] in, Characterizes the deviation value between the sample load-side circuit topology and the load-side circuit topology, The i-th node representing the same topological position of the sample load-side circuit topology and the load-side circuit topology, Characterizes the degree difference between the sample load-side circuit topology and the i-th node of the load-side circuit topology, Represents the branch formed by node x and node y, Characterize the branch set, Characterize the same branch in the topological position of the sample load-side circuit topology and the load-side circuit topology The impedance difference, Characterization weight coefficient, determined based on the Delphi method;

[0183] Construct a function to evaluate the deviation of real-time state information of power grid:

[0184] ,

[0185] in, Characterizes the deviation evaluation value between the real-time state information of the sample power grid and the real-time state information of the power grid, Characterize the k-th attribute grid real-time state parameter, Characterize the normalized deviation of the k-th attribute grid real-time state parameter between the sample grid real-time state information and the grid real-time state information, Characterizes the normalized deviation tolerance value of the predefined k-th attribute grid real-time state parameter, Characterize the preset weight of the k-th attribute power grid real-time state parameter, The total number of parameter attributes representing the real-time status;

[0186] Construct a function to evaluate the deviation of the power distribution equipment tag number:

[0187] ,

[0188] ,

[0189] in, Characterizes the deviation evaluation value between the sample distribution equipment number and the distribution equipment number, The model representing the tag number of the sample power distribution equipment, The model that represents the position number of the power distribution equipment, Characterize the sample distribution equipment number, Characterizes the position number of the power distribution equipment, Characterizes the normalized value of the service time deviation between the sample distribution equipment number and the distribution equipment number, Characterizes the error tolerance value of the normalized value of service time deviation, Characterize the topological distribution location of the sample power distribution equipment number, Characterize the topological distribution location of the power distribution equipment number, Characterizes the Euclidean distance between the sample distribution equipment position number and the topological distribution position of the distribution equipment position number, Characterizes the Euclidean distance deviation threshold, 、 and Characterization weight coefficient, determined based on the Delphi method;

[0190] Construct the demand control modal deviation evaluation function:

[0191] ,

[0192] in, Characterizes the deviation evaluation value between the sample demand control mode and the demand control mode, Characterizes the control mode phase angle deviation between the sample demand control mode and the demand control mode, Characterizes the control parameter vector deviation, Characterization weight coefficient, determined based on the Delphi method;

[0193] Constructing the load pre-connection constraint condition according to the load-side circuit topology deviation evaluation function and the first deviation threshold, and the grid real-time state information deviation evaluation function and the second deviation threshold;

[0194] The load access constraint condition is constructed based on the power distribution equipment bit number deviation evaluation function and the third deviation threshold, and the demand control modal deviation evaluation function and the fourth deviation threshold.

[0195] Furthermore, the centroid of the voltage fluctuation trend of the power distribution grid sample set is extracted to obtain the voltage fluctuation trend of the same-mode power grid, and the execution steps include:

[0196] Clustering the second power grid distribution sample set according to the first power grid distribution sample set to obtain multiple clusters of power grid distribution samples;

[0197] Traversing the plurality of clusters of power distribution grid samples to extract the LOF outlier factor minimum voltage fluctuation trend, and setting it as a plurality of first centroid voltage fluctuation trends;

[0198] extracting a plurality of second voltage fluctuation trends from the first power grid distribution sample set;

[0199] LOF outlier factor minimum voltage fluctuation trend extraction is performed on the multiple first centroid voltage fluctuation trends and the multiple second voltage fluctuation trends to obtain the same-mode power grid voltage fluctuation trend.

[0200] Furthermore, the voltage fluctuation trend of the same-mode power grid is statistically analyzed, and voltage regulation control optimization is performed to obtain a first voltage regulation control parameter. The execution steps include:

[0201] When the voltage variance of the voltage fluctuation trend of the same-mode power grid is greater than or equal to a voltage variance threshold, performing uniform distribution based on the rated interval of the voltage regulation control parameter to obtain a plurality of voltage regulation control parameters;

[0202] By using a voltage regulation simulator bound to a grid access node, the plurality of voltage regulation control parameters are traversed and processed in combination with the same-mode grid voltage fluctuation trend to obtain a plurality of grid voltage fluctuation correction trends;

[0203] When extracting a grid voltage fluctuation correction trend whose voltage variance is less than the voltage variance threshold value from the plurality of grid voltage fluctuation correction trends, the trend is set as the first voltage regulation control parameter.

[0204] Furthermore, the execution step also includes:

[0205] When the number of grid voltage fluctuation correction trends whose voltage variance is less than the voltage variance threshold is equal to 0, sorting the voltage regulation control parameters in ascending order of voltage variance to obtain a voltage regulation control parameter sequence;

[0206] Extracting the first three voltage regulation control parameters of the voltage regulation control parameter sequence and performing same-attribute mean configuration to obtain target voltage regulation control parameters;

[0207] Taking the target voltage regulation control parameter as the regulation target, performing Euclidean distance reduction adjustment on the last 50% of the voltage regulation control parameters in the voltage regulation control parameter sequence to obtain updated voltage regulation control parameters;

[0208] When the voltage variance of the updated voltage regulation control parameter is smaller than the grid voltage fluctuation correction trend of the voltage variance threshold, the first voltage regulation control parameter is set; otherwise, the iteration is continued.

[0209] Furthermore, by using a voltage regulation simulator bound to the grid access node, the plurality of voltage regulation control parameters are traversed and processed in combination with the same-mode grid voltage fluctuation trend to obtain a plurality of grid voltage fluctuation correction trends. The execution step further includes:

[0210] Collecting voltage regulation control logs of the grid access node based on the voltage regulator model, wherein the voltage regulation control logs include initial grid voltage fluctuation trend record data, voltage regulation control parameter record data, and post-regulation grid voltage fluctuation trend record data;

[0211] Configuring a voltage regulation simulator architecture, wherein the voltage regulation simulator architecture includes a first input branch based on a bidirectional gated recurrent network, a second input branch based on a fully connected layer, a backbone network composed of a feature concatenation layer and a temporal convolutional network connected in sequence, and an output layer based on a time-distributed fully connected layer;

[0212] The voltage regulation simulator is trained using the recorded data of the voltage fluctuation trend of the adjusted power grid as supervision and the recorded data of the initial power grid voltage fluctuation trend and the recorded data of the voltage regulation control parameters as input.

[0213] Furthermore, when extracting a grid voltage fluctuation correction trend having a voltage variance less than the voltage variance threshold from the plurality of grid voltage fluctuation correction trends, setting the trend as the first voltage regulation control parameter, the execution steps include:

[0214] When the voltage variance of the updated voltage regulation control parameter is less than the grid voltage fluctuation correction trend of the voltage variance threshold, adding the voltage regulation control parameter to be selected;

[0215] The candidate voltage regulation control parameters are sorted based on minimum energy consumption to obtain the first voltage regulation control parameter.

[0216] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0217] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, approaches, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0218] The present invention is described with reference to flowcharts and / or block diagrams of methods, apparatus (methods), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, 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 computer, or other programmable data processing device to produce 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 flowcharts and / or block diagrams. 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.

[0219] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work 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 The function specified in one or more boxes.

[0220] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0221] Although preferred embodiments of the present invention have been described, additional changes and modifications to these embodiments may occur to those skilled in the art once the basic inventive concepts become known.

[0222] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A voltage regulation power supply control method for a multifunctional power supply, characterized in that: Applicable to a multifunctional power supply voltage regulating power supply device, the multifunctional power supply voltage regulating power supply device includes a grid power supply access circuit and a UPS power supply access circuit, including: Receive a power distribution request uploaded by a user, wherein the power distribution request includes a power distribution equipment position number and a demand control mode; When the power grid is not disconnected, real-time grid status information is obtained, and based on the power distribution equipment position number and the demand control mode, the voltage fluctuation trend of the same-mode grid is statistically analyzed, and voltage regulation control optimization is performed to obtain a first voltage regulation control parameter, and the grid power supply access circuit is started to access the grid and execute the power distribution request based on the first voltage regulation control parameter; When the power grid is outage, the UPS power supply dispatched load status is obtained, and the voltage fluctuation trend of the UPS power supply in the same mode is counted in combination with the power distribution equipment position number and the demand control mode, and the voltage regulation control is optimized to obtain the second voltage regulation control parameter, and the UPS power supply access circuit is started to access the UPS power supply to execute the power distribution request based on the second voltage regulation control parameter; The voltage fluctuation trend of the same-mode power grid is statistically analyzed, and voltage regulation control optimization is performed to obtain the first voltage regulation control parameter, including: When the voltage variance of the voltage fluctuation trend of the same-mode power grid is greater than or equal to a voltage variance threshold, performing uniform distribution based on the rated interval of the voltage regulation control parameter to obtain a plurality of voltage regulation control parameters; By using a voltage regulation simulator bound to a grid access node, the plurality of voltage regulation control parameters are traversed and processed in combination with the same-mode grid voltage fluctuation trend to obtain a plurality of grid voltage fluctuation correction trends; extracting a grid voltage fluctuation correction trend having a voltage variance less than the voltage variance threshold value from the plurality of grid voltage fluctuation correction trends and setting the extracted trend as the first voltage regulation control parameter; When the number of grid voltage fluctuation correction trends whose voltage variance is less than the voltage variance threshold is equal to 0, sorting the voltage regulation control parameters in ascending order of voltage variance to obtain a voltage regulation control parameter sequence; Extracting the first three voltage regulation control parameters of the voltage regulation control parameter sequence and performing same-attribute mean configuration to obtain target voltage regulation control parameters; Taking the target voltage regulation control parameter as the regulation target, performing Euclidean distance reduction adjustment on the last 50% of the voltage regulation control parameters in the voltage regulation control parameter sequence to obtain updated voltage regulation control parameters; When the voltage variance of the updated voltage regulation control parameter is less than the grid voltage fluctuation correction trend of the voltage variance threshold, the first voltage regulation control parameter is set; otherwise, the iteration is continued.

2. The method according to claim 1, wherein When the power grid is not disconnected, obtain real-time status information of the power grid, combine the position number of the power distribution equipment and the demand control mode, and calculate the voltage fluctuation trend of the power grid in the same mode, including: Obtaining a load-side circuit topology including the position number of the power distribution equipment; The load-side circuit topology and the real-time state information of the power grid are used as load pre-connection constraint conditions, and the power distribution equipment position number and the demand control mode are used as load connection constraint conditions; Retrieve a grid distribution sample set that satisfies the load pre-connection constraint condition and the load connection constraint condition through the Internet; The centroid of the voltage fluctuation trend of the power distribution grid sample set is extracted to obtain the voltage fluctuation trend of the same-mode power grid.

3. The method according to claim 2, wherein The network is used to search for a power grid distribution sample set that satisfies the load pre-connection constraint condition and the load connection constraint condition, including: Retrieving a first power grid distribution sample set that satisfies the load pre-connection constraint condition and the load connection constraint condition, wherein any power grid distribution sample in the first power grid distribution sample set includes first power grid real-time state information and a first demand control mode; updating the load pre-connection constraint condition based on the first power grid real-time state information, updating the load access constraint condition based on the first demand control mode, and retrieving a second power grid distribution sample set that satisfies the updated load pre-connection constraint condition and the updated load access constraint condition; The first power grid power distribution sample set and the second power grid power distribution sample set are added to the power grid power distribution sample set.

4. The method according to claim 2, wherein The load-side circuit topology and the real-time state information of the power grid are used as load pre-connection constraint conditions, and the power distribution equipment position number and the demand control mode are used as load connection constraint conditions, including: Construct a load-side circuit topology deviation evaluation function: ， in, Characterizes the deviation between the sample load-side circuit topology and the load-side circuit topology, The i-th node representing the same topological position of the sample load-side circuit topology and the load-side circuit topology, Characterizes the degree difference between the sample load-side circuit topology and the i-th node of the load-side circuit topology, Represents the branch formed by node x and node y, Characterize the branch set, Characterize the same branch in the topological position of the sample load-side circuit topology and the load-side circuit topology The impedance difference, Characterization weight coefficient, determined based on the Delphi method; Construct a function to evaluate the deviation of real-time state information of power grid: ， in, Characterizes the deviation evaluation value between the real-time state information of the sample power grid and the real-time state information of the power grid, Characterize the k-th attribute grid real-time state parameter, Characterize the normalized deviation of the k-th attribute grid real-time state parameter between the sample grid real-time state information and the grid real-time state information, Characterizes the normalized deviation tolerance value of the predefined k-th attribute grid real-time state parameter, Characterize the preset weight of the k-th attribute grid real-time state parameter, The total number of parameter attributes representing the real-time status; Construct a function to evaluate the deviation of the power distribution equipment tag number: , , in, Characterizes the deviation evaluation value between the sample distribution equipment number and the distribution equipment number, The model representing the tag number of the sample power distribution equipment, The model that represents the position number of the power distribution equipment, Characterize the sample distribution equipment number, Characterizes the position number of the power distribution equipment, Characterizes the normalized value of the service time deviation between the sample distribution equipment number and the distribution equipment number, Characterizes the tolerance value of the normalized value of service time deviation, Characterize the topological distribution location of the sample power distribution equipment number, Characterize the topological distribution location of the power distribution equipment number, Characterizes the Euclidean distance between the sample distribution equipment position number and the topological distribution position of the distribution equipment position number, Characterizes the Euclidean distance deviation threshold, 、 and Characterization weight coefficient, determined based on the Delphi method; Construct the demand control modal deviation evaluation function: , in, Characterizes the deviation evaluation value between the sample demand control mode and the demand control mode, Characterizes the control mode phase angle deviation between the sample demand control mode and the demand control mode, Characterizes the control parameter vector deviation, Characterization weight coefficient, determined based on the Delphi method; Constructing the load pre-connection constraint condition according to the load-side circuit topology deviation evaluation function and the first deviation threshold, and the grid real-time state information deviation evaluation function and the second deviation threshold; The load access constraint condition is constructed based on the power distribution equipment bit number deviation evaluation function and the third deviation threshold, and the demand control modal deviation evaluation function and the fourth deviation threshold.

5. The method according to claim 3, wherein Extracting the centroid of the voltage fluctuation trend of the power distribution grid sample set to obtain the voltage fluctuation trend of the same-mode power grid includes: Clustering the second power grid distribution sample set according to the first power grid distribution sample set to obtain multiple clusters of power grid distribution samples; Traversing the plurality of clusters of power distribution grid samples to extract the LOF outlier factor minimum voltage fluctuation trend, and setting it as a plurality of first centroid voltage fluctuation trends; extracting a plurality of second voltage fluctuation trends from the first power grid distribution sample set; LOF outlier factor minimum voltage fluctuation trend extraction is performed on the multiple first centroid voltage fluctuation trends and the multiple second voltage fluctuation trends to obtain the same-mode power grid voltage fluctuation trend.

6. The method according to claim 1, wherein By using a voltage regulation simulator bound to a grid access node, the plurality of voltage regulation control parameters are traversed and processed in combination with the same-mode grid voltage fluctuation trend to obtain a plurality of grid voltage fluctuation correction trends, including: Collecting voltage regulation control logs of the grid access node based on the voltage regulator model, wherein the voltage regulation control logs include initial grid voltage fluctuation trend record data, voltage regulation control parameter record data, and post-regulation grid voltage fluctuation trend record data; Configuring a voltage regulation simulator architecture, wherein the voltage regulation simulator architecture includes a first input branch based on a bidirectional gated recurrent network, a second input branch based on a fully connected layer, a backbone network composed of a feature concatenation layer and a temporal convolutional network connected in sequence, and an output layer based on a time-distributed fully connected layer; The voltage regulation simulator is trained using the recorded data of the voltage fluctuation trend of the adjusted power grid as supervision and the recorded data of the initial power grid voltage fluctuation trend and the recorded data of the voltage regulation control parameters as input.

7. The method according to claim 6, wherein When extracting a grid voltage fluctuation correction trend having a voltage variance less than the voltage variance threshold value from the plurality of grid voltage fluctuation correction trends, setting the trend as the first voltage regulation control parameter includes: When the voltage variance of the updated voltage regulation control parameter is less than the grid voltage fluctuation correction trend of the voltage variance threshold, adding the voltage regulation control parameter to be selected; The candidate voltage regulation control parameters are sorted based on minimum energy consumption to obtain the first voltage regulation control parameter.

8. A voltage regulating power supply control system for a multifunctional power supply, characterized in that: Applicable to a multifunctional power supply voltage regulating power supply device, the multifunctional power supply voltage regulating power supply device includes a grid power supply access circuit and a UPS power supply access circuit, including: A power distribution request receiving module, configured to receive a power distribution request uploaded by a user, wherein the power distribution request includes a power distribution equipment position number and a demand control mode; A first power distribution control module is configured to, when the power grid is not disconnected, obtain real-time power grid status information, combine the power distribution equipment position number and the demand control mode, calculate the voltage fluctuation trend of the same-mode power grid, perform voltage regulation control optimization, obtain a first voltage regulation control parameter, start the power grid power supply access circuit to connect to the power grid, and execute the power distribution request based on the first voltage regulation control parameter; The second power distribution control module is configured to obtain the scheduled load status of the UPS power supply when the power grid is outage, calculate the voltage fluctuation trend of the UPS power supply in the same mode based on the power distribution equipment position number and the demand control mode, perform voltage regulation control optimization, obtain the second voltage regulation control parameter, start the UPS power supply access circuit to access the UPS power supply, and execute the power distribution request based on the second voltage regulation control parameter; The voltage fluctuation trend of the same-mode power grid is statistically analyzed, and voltage regulation control optimization is performed to obtain the first voltage regulation control parameter, including: When the voltage variance of the voltage fluctuation trend of the same-mode power grid is greater than or equal to a voltage variance threshold, performing uniform distribution based on the rated interval of the voltage regulation control parameter to obtain a plurality of voltage regulation control parameters; By using a voltage regulation simulator bound to a grid access node, the plurality of voltage regulation control parameters are traversed and processed in combination with the same-mode grid voltage fluctuation trend to obtain a plurality of grid voltage fluctuation correction trends; extracting a grid voltage fluctuation correction trend having a voltage variance less than the voltage variance threshold value from the plurality of grid voltage fluctuation correction trends and setting the extracted trend as the first voltage regulation control parameter; When the number of grid voltage fluctuation correction trends whose voltage variance is less than the voltage variance threshold is equal to 0, sorting the voltage regulation control parameters in ascending order of voltage variance to obtain a voltage regulation control parameter sequence; Extracting the first three voltage regulation control parameters of the voltage regulation control parameter sequence and performing same-attribute mean configuration to obtain target voltage regulation control parameters; Taking the target voltage regulation control parameter as the regulation target, performing Euclidean distance reduction adjustment on the last 50% of the voltage regulation control parameters in the voltage regulation control parameter sequence to obtain updated voltage regulation control parameters; When the voltage variance of the updated voltage regulation control parameter is less than the grid voltage fluctuation correction trend of the voltage variance threshold, the first voltage regulation control parameter is set; otherwise, the iteration is continued.

Images