Distribution network area voltage treatment method and system

Abstract

The invention provides a voltage treatment method and system for a distribution network area, and the method is applied to a voltage treatment device which is installed in series at a low-voltage side of the distribution network area, and comprises the steps: collecting electrical quantity data and non-electrical quantity data of the area in real time, carrying out the storage and preprocessing of the collected data, and uploading the data to a main station according to a preset rule; on the basis of the acquired electrical quantity and non-electrical quantity data, carrying out transformer area load flow calculation and rapid load flow result correction based on a sensitivity method to obtain voltage of each node of the transformer area; generating a control instruction according to a preset target voltage range, and adjusting device output voltage; in combination with the control instruction, the control execution effect and a device fault self-checking method based on an innovation graph method, the running state of the device is researched and judged; fault self-healing is tried, and fault related information and disposal suggestions are pushed. According to the invention, feasibility and stability of transformer area voltage treatment can be improved.

Description

Technical Field

[0001] This invention belongs to the field of power system automation technology, specifically relating to a method and system for voltage management in distribution network areas. Background Technology

[0002] Low-voltage distribution networks directly face users, forming the "last mile" of power supply and determining the quality of electricity delivered. In rural distribution networks, due to excessively long supply radii and the large-scale integration of distributed photovoltaic (PV) systems, bidirectional voltage exceedance issues are becoming increasingly prominent. The construction of new power supply voltage systems requires strengthening the management of low-voltage problems, guiding the local consumption of distributed energy, and providing targeted solutions for voltage exceedance and PV carrying capacity. Local management of high and low voltage exceedance issues at the distribution level is a trend in voltage management within new power systems. However, the real-time nature and accuracy of distribution area data, as well as insufficient centralized control, monitoring, and calculation methods, all constrain the foundation for distribution area power flow perception. Furthermore, the reliability of distribution area voltage management equipment directly affects the safe and stable operation of the distribution network. Three major problems—difficulty in voltage perception, difficulty in interaction between calculation and management, and difficulty in equipment maintenance—restrict the implementation of voltage management at the distribution area level. Therefore, it is urgent to improve distribution area power flow calculation methods through technical means to adapt them to the existing measurement foundation, and on this basis, improve the online strategy adjustment and self-diagnosis of voltage exceedance management equipment to enhance the feasibility and stability of distribution area voltage management. Summary of the Invention

[0003] The purpose of this application is to provide a method and system for managing voltage in distribution network areas, which can improve the feasibility and stability of voltage management in distribution network areas.

[0004] The technical solution provided in this application is:

[0005] In a first aspect, this application provides a method for voltage regulation in a distribution network area, which is applied to a voltage regulation device installed in series on the low-voltage side of the distribution network area, comprising:

[0006] Data acquisition: Real-time acquisition of electrical and non-electrical quantity data of the transformer area, storage, preprocessing of the acquired data and uploading to the main station according to preset rules;

[0007] Control strategy generation: Based on electrical and non-electrical quantity data, power flow calculation of the transformer area is carried out and fast power flow result correction based on sensitivity method is performed to obtain the voltage of each node in the transformer area; control commands are generated according to the preset target voltage range to adjust the output voltage of the device;

[0008] Fault self-diagnosis and handling: By combining control commands and control execution effects, as well as the device fault self-diagnosis method based on the hologram method, the device operating status is judged; fault self-healing is attempted and fault-related information and handling suggestions are pushed out.

[0009] In the data acquisition stage, the above solution solves the problem of insufficient real-time performance and completeness of traditional transformer area data by collecting and uploading electrical quantities (voltage, current, power factor, etc.) and non-electrical quantities in real time, thus providing a high-quality data foundation for subsequent power flow calculation and fault diagnosis.

[0010] In the control strategy generation stage, power flow calculations are performed on the transformer area based on the collected data. Control commands are generated in conjunction with the preset target voltage range, realizing the transformation of voltage management from "passive response" to "active regulation". This effectively solves the problem of bidirectional voltage over-limit (high voltage, low voltage, voltage imbalance) and ensures power quality on the user side.

[0011] In the fault self-inspection and handling stage: the effect of control command execution is combined with the fault self-inspection method of hologram to realize dynamic judgment of the device's operating status, avoid the faulty device from causing secondary impact on the power grid, and at the same time, through fault self-healing and information push, the reliance on manual maintenance is reduced, and the fault tolerance and reliability of the governance system are improved.

[0012] In one possible implementation, the real-time acquisition of electrical and non-electrical quantity data of the transformer substation, storage, preprocessing, and uploading of the acquired data to the main station according to preset rules includes: real-time acquisition of electrical quantity data (such as voltage, current, power factor, etc.) through electrical quantity detection devices installed on the power supply side and load side of the transformer substation; acquisition of non-electrical quantity data through environmental parameter monitoring devices installed in the area where the transformer substation is located; acquisition, storage, and preprocessing of the data according to preset local acquisition intervals; uploading of the data to the main station according to preset upload intervals; and uploading of all data to the main station according to local acquisition intervals if there is a voltage over-limit problem.

[0013] In one possible implementation, the power flow calculation for the distribution area is based on a typical scenario voltage calculation model for distribution areas with distributed photovoltaic (PV) power. This model divides the distribution area nodes into PQ nodes, PV nodes, and slack nodes, and considers the distribution area voltage mitigation device as an adjustable impedance of the power supply network connected to the distribution area. Specifically, a PQ node refers to a node with a given fixed active power P and reactive power Q, and a constant apparent power S = P + Q; a PV node refers to a node with given P and reactive power U; and a slack node refers to a node with given reactive power U. The calculation process includes:

[0014] 1) Initialization: Set the initial voltage values, initial power values, admittance matrix, and other parameters required for calculation at each node in the transformer area;

[0015] 2) Calculate the initial value of the node current: Calculate according to the formula I(0)=S(0) / U(0), where S(0) is the initial value of the node power and U(0) is the initial value of the node voltage;

[0016] 3) Iteratively calculate the voltage deviation of the node relative to the slack node: using the formula Iterative calculation of voltage deviation ,in Let I(n) be the inverse of the nodal admittance matrix, and let I(n) be the nodal current in the nth iteration. To balance the node voltage;

[0017] 4) Restore PQ node voltage: Based on the balanced node voltage, the voltage deviation obtained through iteration is... Restored to the actual PQ node voltage: ;

[0018] 5) Determine if node voltages have converged: Check if the node voltages meet the convergence criteria (e.g., the voltage changes of all PQ nodes in two consecutive iterations are less than a preset threshold). If converged, then proceed according to... Calculate the power of each node after convergence, and output the voltage and apparent power of each node; if convergence is not achieved, reset the initial value of the PQ node voltage to the latest iteration value, and return to step 2) to continue iterating.

[0019] The above scheme divides the transformer substation into PQ nodes, PV nodes, and balancing nodes, and treats the voltage management device as an adjustable impedance. It accurately matches the different characteristics of distributed photovoltaic (such as inverter power factor priority / voltage priority mode) and user load, avoiding the calculation errors caused by the simplification of node characteristics in traditional models.

[0020] In one possible implementation, if the voltage of the PQ node is found to be exceeding the limit, the node is converted into a PV node and its voltage is set to equal its voltage limit; if the apparent power S=P+Q of the PV node is found to be exceeding the limit, the node is converted into a PQ node and its apparent power S is set to equal its apparent power limit.

[0021] This scheme automatically switches PQ nodes to PV nodes (voltage controlled by limits) when the voltage exceeds the limit, and automatically switches PV nodes to PQ nodes (apparent power controlled by limits) when the apparent power (injected power) exceeds the limit. This solves the problem of dynamic changes in node state caused by fluctuations in distributed photovoltaic output, ensuring that power flow calculations can still converge and produce accurate results under complex operating conditions.

[0022] The above scheme reduces the complexity of power flow calculation by ignoring small power angles and simplifying AC vectors into scalars, enabling rapid iteration of voltage and power at each node in the transformer area and providing real-time data support for the generation of subsequent control strategies.

[0023] In one possible implementation, the control strategy generation step further includes: if the grid voltage is normal, a device pass-through strategy is adopted; if the voltage exceeds the limit, the device is put into operation; and if the device fails, the device is short-circuited through a short-circuit component.

[0024] When the grid voltage in the distribution area is within the preset normal range, the voltage management device does not need to activate its regulation function. A direct-through strategy allows the grid current to pass directly through the device (without passing through an additional voltage regulation module), avoiding ineffective operation when there is no need for regulation. When the distribution area experiences overvoltage, undervoltage, or voltage imbalance issues, the device immediately switches from direct-through mode to operational mode. Based on the voltage data of each node obtained from prior power flow calculations and sensitivity correction, it generates precise control commands, efficiently responding to management needs and ensuring power quality. When the voltage management device itself malfunctions (such as thyristor breakdown or regulation module failure), the system triggers a short-circuit component (such as a double reverse-connected parallel thyristor) to short-circuit the device, forming a backup path for the grid current, ensuring continuous power supply and mitigating the risk of fault propagation.

[0025] In one possible implementation, the fast power flow result correction based on the sensitivity method includes the following steps:

[0026] Input the typical operating data set of the transformer substation corresponding to the current scenario. The data set includes the active power, reactive power, and voltage values ​​of the load nodes and the active power, reactive power, and voltage values ​​of the distributed photovoltaic nodes under different seasons, weather conditions, and time periods.

[0027] The status changes of each monitoring point in the monitoring area ;

[0028] Determine whether the detected state change originates from the power output fluctuation of the photovoltaic monitoring point. If so, supplement the active and reactive power of the photovoltaic at the non-monitoring point through state estimation; otherwise, proceed directly to the next step.

[0029] judge If the value is less than the preset threshold α, the correction amount is calculated using the sensitivity method: by establishing the sensitivity relationship between the state variable and the control variable, the parameter correction value that needs to be adjusted due to the state change is quickly calculated; if not, a complete power flow iteration calculation is performed and the typical scenario data set is updated.

[0030] Output power flow distribution status, which includes the voltage and power of each node.

[0031] This scheme, based on the sensitivity method for rapid power flow result correction, overcomes the pain points of traditional complete power flow iteration calculations being time-consuming and resource-intensive. By pre-storing transformer operation data sets for different seasons, weather conditions, and time periods, it avoids the need to perform a complete iteration for each state change, enabling the reuse of typical data sets and significantly reducing computational resource consumption. It supplements the photovoltaic power of non-monitoring points through state estimation, and only when the state change... A full iteration is initiated only when the threshold α is exceeded; otherwise, the correction amount is quickly calculated using the sensitivity method. This solves the problem of limited measurement nodes in the distribution area and achieves dynamic adaptation of "rapid correction for small fluctuations and accurate calculation for large fluctuations," ensuring that power flow calculations maintain both real-time performance and accuracy even when photovoltaic output fluctuates drastically. This correction scheme can be deployed at the edge nodes of the distribution transformer box in the distribution area without relying on the large-scale computing power of the main station, reducing investment in communication and computing resources and adapting to the infrastructure conditions of rural power distribution networks.

[0032] In one possible implementation, the device fault self-testing method based on the neohologram method includes the following steps:

[0033] Calculate the innovation vector: Innovation vector = Current measurement vector - Current forecast vector, where the current forecast vector is calculated from the power flow at the previous time, and the current measurement vector is obtained from the actual measurement value at the current time;

[0034] Using the low-voltage side of the distribution transformer as the root node, each line in the transformer area as a tree branch, and the nodes in the transformer area as connecting branches, a new information vector diagram of the transformer area is constructed.

[0035] Calculate the cumulative innovation vector of the path: From the root node to the terminal load node, calculate the cumulative innovation vector of the path according to the formula. If the cumulative innovation vector of the path is greater than the preset threshold, it is determined that there is bad data in the system and suspicious branch identification is carried out.

[0036] Reverse calculation of tree branch information: Perform reverse calculation of tree branch information from the terminal node to the root node, calculate the node load information, branch load information, expected branch load information and information difference respectively, where the information difference is the difference between the current information vector of the branch and the expected branch load information;

[0037] Fault diagnosis: The new interest difference of the branch in the transformer area is formed into a new interest difference vector. The non-zero elements in the new interest difference vector are sorted according to the branch connection relationship to form an abnormal path starting from the root node of the transformer area. The fault of the transformer area voltage management device in the corresponding branch is determined based on the extension range of the abnormal path.

[0038] This solution proposes a novel information map method for device fault self-diagnosis, solving the problems of traditional fault detection relying on manual investigation and low location accuracy. It calculates the information vector by subtracting the predicted vector from the current measurement vector, and constructs an information vector diagram with the low-voltage side of the distribution transformer as the root node, lines as branches, and nodes as connections. This associates abstract electrical quantity errors with the physical topology, avoiding misjudgments caused by the disconnect between data and topology in traditional fault detection. The existence of bad data is determined by accumulating information along the path from the root node to the end node. Then, through reverse calculation from the end node to the root node (calculating node / branch load information, expected information, and information difference), abnormal paths and faulty branches are accurately located, solving the problem of difficulty in quickly locating fault points in the radial distribution network. Device faults are equated to erroneous distribution network parameters. By sorting the paths of non-zero elements in the information difference vector, faulty branches and voltage management devices are directly associated, achieving an automatic fault identification-precise location-information push closed loop, significantly shortening fault handling time.

[0039] In one possible implementation, the fault self-check and handling steps include attempting fault self-healing and pushing fault-related information and handling suggestions, including taking maintenance-free measures to attempt fault self-healing, such as adaptive adjustment of control parameters or restarting, etc. If self-healing fails, fault information, measures already taken and manual handling suggestions are pushed to the operation and maintenance personnel, and a defect report is generated by uploading it to the main station.

[0040] In one possible implementation, the method further includes:

[0041] Remote Interaction: Through a remote interaction module based on the MQTT protocol, the device uploads local information to the master station and receives the collaborative control strategy given by the master station based on the power flow calculation of the distribution network, realizing the collaborative control of multiple devices or the optimization of control parameters in the mode of human intervention; the device connects to the intelligent converged terminal of the distribution area through wired (such as fiber optic) or wireless communication, and realizes data transmission, device status visualization and real-time local control based on the preset communication protocol (4G communication is used for the physical layer and link layer, TCP protocol is used for the network layer and transport layer, and MQTT protocol is used for the application layer).

[0042] This solution, based on the MQTT protocol, constructs a remote interaction module, solving the problems of isolated operation and weak coordination capabilities of traditional distribution voltage management devices. By generating coordination strategies through power flow calculations on the master station side, it enables the coordinated control of multiple voltage management devices, avoiding uneven voltage distribution in the distribution area caused by single-device control. This is particularly suitable for complex distribution areas with large-scale distributed photovoltaic access. It supports fiber optic and wireless (4G) communication, and adopts standardized protocols (4G+TCP+MQTT) at the physical layer, link layer, network layer, and application layer, allowing for flexible adaptation of communication methods. This solves the problem of significant differences in communication infrastructure in rural distribution networks, ensuring stable access of devices to the distribution area's intelligent converged terminal. It supports control parameter optimization in manned mode, retaining the efficiency of automated control while providing flexible intervention interfaces for maintenance personnel, adapting to different management needs in various scenarios (such as manual emergency handling of sudden faults and customized voltage curves for special periods).

[0043] Secondly, this application provides a distribution network area voltage management system, comprising:

[0044] Data acquisition module: used to collect electrical and non-electrical data of the transformer area in real time, store the collected data, preprocess the data, and upload it to the main station according to preset rules;

[0045] Control strategy generation module: used to perform power flow calculations for transformer substations based on electrical and non-electrical quantity data and to correct the rapid power flow results based on the sensitivity method, thereby obtaining the voltage of each node in the transformer substation; and to generate control commands according to the preset target voltage range to adjust the output voltage of the device.

[0046] Fault self-diagnosis and handling module: It is used to analyze the device's operating status by combining control commands and control execution effects, as well as the device fault self-diagnosis method based on the hologram method; attempt fault self-healing; and push fault-related information and handling suggestions.

[0047] The voltage management system for distribution network areas uses the above-described method to manage voltage in distribution network areas.

[0048] Thirdly, this application provides an electronic device, including: a memory and a processor;

[0049] The memory is used to store computer programs;

[0050] The processor is used to invoke the computer program to execute the method described above.

[0051] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed on an electronic device, causes the electronic device to perform the method described above.

[0052] Fifthly, this application provides a computer program product, including a computer program that, when run on an electronic device, causes the electronic device to perform the method described above.

[0053] The specific implementation methods of the second to fifth aspects of this application can refer to the implementation methods of the first aspect, and will not be elaborated here.

[0054] This application has the following beneficial effects:

[0055] Improved governance accuracy: Reduced power flow calculation errors in scenarios involving distributed photovoltaics, increased the success rate of voltage limit violation governance, and improved power quality on the user side;

[0056] Improved operational efficiency: Sensitivity correction and neostructure method for fault self-diagnosis significantly shorten calculation and fault handling time; multi-device collaboration and remote interaction reduce manual intervention costs.

[0057] Improved economic efficiency: By using edge computing and reusing typical data sets, investment in computing power and communication resources is reduced. Fault self-healing reduces equipment replacement and maintenance costs, providing a cost-effective solution for voltage management in rural power distribution networks. Attached Figure Description

[0058] Figure 1 This is a functional structure diagram of a voltage-friendly transformer substation voltage management device in one embodiment of this application;

[0059] Figure 2 This is an equivalent circuit for a distributed photovoltaic substation in one embodiment of this application;

[0060] Figure 3 This is a flowchart of photovoltaic power flow calculation in one embodiment of this application;

[0061] Figure 4 This is a flowchart illustrating the rapid power flow correction process for a photovoltaic power distribution area in one embodiment of this application.

[0062] Figure 5 This is a schematic diagram of the topology of the substation in one embodiment of this application. Detailed Implementation

[0063] To enable those skilled in the art to better understand the present application, the technical solution of the present application will be further described in detail below with reference to the embodiments and accompanying drawings.

[0064] This application provides a method and system for voltage management in distribution network areas, achieving voltage management based on agile control and intelligent self-testing. Agile control refers to the ability of relevant equipment to quickly and in real-time generate control schemes based on limited data, continuously supporting the improvement of low-voltage side voltage quality in the distribution area, which is a crucial guarantee for distributed photovoltaic grid integration. This invention revolves around a grid-friendly voltage management device (voltage limit management device). Based on limited measurement and communication in the distribution network, it efficiently interacts with the grid, constructing a control platform with functions such as rapid power flow calculation, control strategy generation, and fault / anomaly self-detection. This enables the voltage management device to simultaneously solve multiple voltage limit problems such as low voltage, high voltage, and voltage imbalance. The system consists of four main parts: a high-precision data acquisition module, a remote interaction module based on the MQTT protocol, a control strategy generation module, and a fault / anomaly self-testing and handling module. By introducing a grid-friendly concept, it specifically improves the shortcomings of traditional voltage management devices, comprehensively enhancing the efficiency of low-voltage side voltage monitoring and management in the distribution area, and achieving efficient and reliable operation of the device.

[0065] Specific embodiments according to this application will now be described with reference to the accompanying drawings.

[0066] Example 1:

[0067] The functional structure diagram of the following method is shown below. Figure 1 As shown.

[0068] This application provides a method for voltage regulation in a distribution network area, which is applied to a voltage regulation device installed in series on the low-voltage side of the distribution network area, including:

[0069] Data acquisition: Real-time acquisition of electrical and non-electrical quantity data of the transformer area, storage, preprocessing of the acquired data and uploading to the main station according to preset rules;

[0070] Control strategy generation: Based on electrical and non-electrical quantity data, power flow calculation of the transformer area is carried out and fast power flow result correction based on sensitivity method is performed to obtain the voltage of each node in the transformer area; control commands are generated according to the preset target voltage range to adjust the output voltage of the device;

[0071] Fault self-diagnosis and handling: By combining control commands and control execution effects, as well as the device fault self-diagnosis method based on the hologram method, the device operating status is judged; fault self-healing is attempted, and fault-related information and handling suggestions are pushed out.

[0072] In some embodiments, the voltage regulation device is installed in series on the low-voltage side of a designated distribution network area. Voltage transformers, current transformers, and other detection devices are installed on the power supply and load sides of the area to collect electrical quantity data such as voltage, current, and power factor from upstream and downstream nodes in real time. Temperature monitoring devices are installed in the area where the area is located to collect non-electrical quantity data such as temperature for subsequent analysis and control. Local data is collected, stored, and preprocessed at a rate of 1 minute per point, and relevant data is uploaded to the master station every 15 minutes. If voltage exceedance issues (high voltage, low voltage, voltage imbalance) exist, all data collected every minute is uploaded to the master station.

[0073] In some embodiments, the device's local information is transmitted to the master station via a remote interaction system. The master station then transmits the collaborative control strategy based on power flow calculations to the device, enabling collaborative control of multiple devices or optimized control parameters in a manually intervened mode. This remote operation capability allows for rapid and precise voltage adjustment without manual intervention. The voltage management device connects to the nearest intelligent converged terminal in the distribution area via fiber optic or wireless public networks. 4G communication is used at the physical and link layers, and TCP protocol at the network and transport layers for data transmission and reception. The MQTT protocol is used at the application layer to achieve status visualization and real-time local control of the voltage management device.

[0074] In some embodiments, the control strategy generation stage is the core of the voltage regulation device. It receives electrical and non-electrical data from the data acquisition module, acts as an edge computing node to perform power flow calculations for the transformer substations, and performs rapid power flow result correction based on the sensitivity method to obtain the voltage of each node in the substation area. It then compares and judges the voltage according to the preset target voltage range of the voltage regulation device, generates corresponding control commands, and adjusts the output voltage of the device to achieve grid voltage regulation and stability. Under normal grid voltage conditions, the voltage regulation device adopts a direct-on control strategy to reduce device losses and the possibility of device failure. When the grid voltage is too high or too low, the voltage regulation device is activated. In the event of a voltage regulation device failure, a double reverse-connected parallel thyristor is used to short-circuit the device, without affecting the normal power supply of the grid.

[0075] In some embodiments, the fault anomaly self-checking and handling module performs two main functions: first, it analyzes and judges the device's operating status by combining the control commands generated by the control strategy generation module and the control execution effects collected through subsequent tracking; second, it analyzes the device's abnormal operating conditions using the fault self-checking method for the transformer voltage management device described in this application. By taking maintenance-free measures such as adaptive adjustment of the faulty device's control parameters or restarting it, the module determines the possibility of the faulty device self-healing. If self-healing is not possible, it automatically pushes relevant fault information, measures already taken, and subsequent manual handling suggestions to relevant maintenance personnel, and uploads the above information to the remote master station to generate a defect report.

[0076] The algorithm in the embodiments of this application will be analyzed and explained in detail below.

[0077] (I) Voltage calculation model for typical scenarios including distributed photovoltaic systems:

[0078] Figure 2 The diagram shows the equivalent circuit of a distributed photovoltaic (PV) distribution area. Compared to high-voltage transmission lines, the 400V distribution area line has a smaller conductor cross-sectional area, a much larger line resistance R than line reactance X, and a smaller power angle. Therefore, the AC vector can be approximated as a scalar. The original line impedance Z... 0 =R+jX and complex power S 0 =P+jQ, after realization, we get Z=R+X, S=P+Q; where Z and S represent the line impedance and apparent power after realization, respectively, j is the imaginary unit, P is the active power, and Q is the reactive power.

[0079] The specific calculation model for transformer substation voltage is as follows:

[0080] 1) PQ node: refers to the node where, given a fixed active power P and reactive power Q, and the apparent power S = P + Q is constant, the voltage U and phase angle δ are solved.

[0081] If the calculated voltage U of the node exceeds the limit, the node is converted into a PV node, and the voltage is equal to the voltage limit of the node.

[0082] When the distributed photovoltaic (PV) inverter is set to power factor priority mode, the distributed PV can be considered as a PQ node. User pure load nodes can also be considered as PQ nodes.

[0083] 2) PV node: That is, voltage control node, which refers to the node where P and U are given and Q and voltage phase angle δ are solved; if the apparent power S=P+Q of the node is exceeded, the node is transformed into a PQ node and the apparent power S is equal to the apparent power limit of the node.

[0084] When the distributed photovoltaic inverter is in voltage priority mode, the inverter has a certain reactive power regulation capability and is regarded as a PV node; if the reactive power exceeds the limit, it is converted into a PQ node.

[0085] 3) Balance node: refers to the node where P and Q are solved given U and δ (δ=0° reference), that is, the node in the distribution area where the voltage always remains constant; generally it is the 10 kV bus of the substation or the voltage side of the distribution transformer in the distribution area.

[0086] 4) Distribution transformer voltage management device; an adjustable impedance in the power supply network connected to the distribution transformer area. It serves as a control variable for adjusting the voltage distribution in the distribution transformer area.

[0087] By inputting the parameters of each branch of the transformer area, the admittance matrix Y of the transformer area nodes and its inverse matrix Y can be obtained.-1 The power of the PQ node and the voltage values ​​of the PV node and the slack node are known inputs. By setting the initial voltage values ​​of the PQ node and the initial power values ​​of the PV node, the complete initial voltage value U(0) and initial power value S(0) for each node are obtained. Specifically, after adopting the real-valued method, the injected current I=S / U and the power flow equations based on the node admittance matrix for each node are obtained. That is, it becomes a real number-based numerical calculation.

[0088] According to Figure 3 The steps shown can quickly perform power flow calculations to obtain the voltage of each node in the transformer area, including the following steps:

[0089] 1) Initialization: Set the initial parameters required for the calculation (such as initial values ​​of node voltages, admittance matrix, etc.).

[0090] 2) Calculate the initial current of each node: Calculate the initial current of each node according to the formula I(0)=S(0) / U(0), where S(0) is the initial power of the node and U(0) is the initial voltage of the node.

[0091] 3) Iterate the voltage of each node relative to the equilibrium node: using the formula Iteratively calculate the voltage of each node relative to the slack node, where Y -1 It is the inverse of the node admittance matrix, I(n) is the node current in the nth iteration, and U(n) is the equilibrium node voltage in the nth iteration.

[0092] 4) Restore PQ node voltage: Restore the voltage deviation ∆U obtained from the iteration to the actual PQ node voltage.

[0093] 5) Determine if node voltages have converged: Check if the node voltages meet the convergence condition (i.e., the voltage change between two adjacent iterations is less than a preset threshold). If converged, output the voltage of each node and apply the condition according to S= Calculate the power of each node after convergence. If convergence fails, reset the initial voltage value of node PQ to the latest iteration value and return to step 2) to continue iteration.

[0094] Based on the above power flow calculation model for transformer areas, scenarios are generated for different seasons, weather conditions (sunny, cloudy, rainy), and time periods. Typical voltage distribution data set of the following distribution area ,in These represent the active power, reactive power, and voltage values ​​of load node n in scenario i, respectively. Let n be the active power, reactive power, and voltage values ​​of distributed photovoltaic node n in scenario i.

[0095] (II) Edge Computation Perturbation Analysis Based on Sensitivity Method

[0096] The large-scale integration of distributed photovoltaic (PV) systems, due to their unstable power output, leads to voltage fluctuations and other issues, placing high demands on the real-time performance of power flow calculations. Furthermore, the regulation strategy formulation for transformer substation voltage control devices also relies on power flow calculations. However, the limited number of measurement points in transformer substations, coupled with their low real-time performance, makes it difficult to meet the requirements of substation voltage regulation. This application deploys edge computing nodes in the transformer substation boxes and designs a reasonable fast power flow correction strategy by combining photovoltaic power output state estimation and sensitivity calculation. This strategy aims to meet the needs of substation regulation while minimizing the number of power flow calculations to save computational costs, thereby enabling real-time monitoring and remote control of substation voltage. The process is as follows: Figure 4 As shown. The specific steps include:

[0097] 1) Input a typical data set for this scenario;

[0098] For example, import typical operating data of the transformer substation corresponding to the current scenario (such as different seasons, weather, and time periods), including basic parameters such as power and voltage of each node.

[0099] 2) Status changes at each monitoring point in the monitoring area It is used to determine the degree of fluctuation in the system state;

[0100] 3) Determine whether the monitored state changes originate from power output fluctuations at the photovoltaic monitoring point;

[0101] If so, the active and reactive power outputs of photovoltaics at non-monitoring points are supplemented by state estimation, that is, the power output of photovoltaic nodes that are not directly monitored is calculated using the photovoltaic data of the monitored points.

[0102] If not, proceed directly to the next step;

[0103] 4) Judgment Is it less than the preset threshold α?

[0104] <α indicates whether the system state fluctuations are within the range that can be quickly corrected using the sensitivity method;

[0105] If so, the sensitivity method is used to calculate the correction amount, that is, by establishing the sensitivity relationship between the state variable and the control variable, the parameter correction value that needs to be adjusted due to the change of state is quickly calculated.

[0106] If not, then power flow calculation is performed, which involves obtaining the voltage and power distribution of each node in the transformer area through a complete power flow iteration process (such as the power flow calculation process mentioned above), and then updating the typical scenario data set to provide new basic data for subsequent calculations.

[0107] 5) Output the final power flow distribution status of the transformer area, including key information such as voltage and power of each node.

[0108] The sensitivity method is calculated based on the power flow equation, which is described as follows:

[0109] (1)

[0110] These are the system's state variables, mainly the voltage and phase at each measurement point in the distribution area; These are the control variables of the system, mainly the photovoltaic output and the equivalent impedance of the voltage management device in the distribution area. The system's power constraint equations and area power flow balance equations represent the equations under given state variables. and control variables When the active and reactive power of the transformer area reach equilibrium, it is the core constraint condition for power flow calculation. For the output equation, describe the output variables. With state variables and control variables Functional relationship between them; output variables These are dependent variables related to the aforementioned control and state variables, such as line active power and network losses.

[0111] Through At this point, performing a Taylor expansion and neglecting terms of degree two and above, we obtain:

[0112] (2)

[0113] Substituting equation (1) into equation (2):

[0114] (3)

[0115] We can obtain:

[0116] (4)

[0117] In the formula, It represents the change in state variables (state change), that is, the change in state variables (such as voltage and phase) caused by factors such as fluctuations in photovoltaic power output; It represents the amount of change in the controlled variable, such as the amount of change in photovoltaic output or the amount of adjustment of the equivalent impedance of the voltage regulation device; This is the sensitivity matrix of the state variable to the control variable, which describes the degree of change of the state variable when the control variable changes by one unit. This is the sensitivity matrix of the output variable to the control variable, which describes the degree of change in the output variable when the control variable changes by one unit.

[0118] Equation (4) can be used to quickly calculate the voltage distribution under photovoltaic power output fluctuations. By optimizing the control parameters of the voltage management device in the distribution area and changing the equivalent impedance, the voltage of each node in the distribution area can be controlled within the acceptable range.

[0119] (III) Self-inspection method for control faults of voltage management devices in distribution areas

[0120] When a voltage management device in a distribution area experiences a fault such as thyristor breakdown, its conduction performance changes. The adjustable impedance in the power supply network connected to the distribution area becomes inconsistent with the adjustable impedance corresponding to the set parameters. This is treated as erroneous data of the distribution network parameters.

[0121] This application proposes a device fault self-checking method based on the neologism method, grounded in neologism vectors and graph theory. The neologism vector is obtained by subtracting the predicted vector from the measured vector, representing the error between prediction and measurement, and includes information such as bad data. The distribution of the neologism vector is analyzed by combining the distribution network topology and the neologism vector map to screen for device faults. The predicted vector at the current moment is obtained from the power flow meter at the previous moment (see sections (I) and (II) of this application for details), and the measured vector at the current moment is obtained from the actual measured value at the current moment.

[0122] Equation (5) gives the expression for the active information vector.

[0123] (5)

[0124] In the formula, For the current information vector, Measure vectors at the current time. This is the forecast vector for the current time.

[0125] Using the above formula, the information vector of each node and branch in the transformer substation can be calculated. An information vector diagram based on the transformer substation is then constructed. The transformer outlet side is the root node. Because the distribution network operates in an open loop, each line in the substation is a branch, and the substation node is equivalent to a connecting branch. A schematic diagram of the substation topology is shown below. Figure 5 As shown.

[0126] Calculate the path cumulative innovation vector from the root node to the terminal load node. :

[0127] (6)

[0128] In the formula, n is the set of nodes belonging to the path road from the root node to the terminal node 3 / 5. The nodes in Let n be the current information vector of node n. If the path cumulative information vector... If the value exceeds the threshold, it indicates the presence of bad data in the system, and suspicious branch identification will be carried out.

[0129] The tree branch information analysis is performed by reversing the branch from the end node to the root node.

[0130] The load information for node n and branch l are as follows:

[0131] (7)

[0132] (8)

[0133] In the formula, Let n be the current information vector. Measure the vector at the current time for node n. This is the current time-based prediction vector for node n; branch road Current moment's new information vector, branch road Current measurement vector, branch road Forecast vector at the current moment.

[0134] The expected branch load information for branch l is as follows:

[0135] (9)

[0136] in, branch road The expected branch load information, where k is the branch. downstream nodes, Indicates a branch The set of downstream nodes.

[0137] The difference between the current innovation vector (actual innovation) of branch l and the expected branch load innovation is defined as the innovation difference:

[0138] (10)

[0139] In the formula, branch road The new interest rate spread;

[0140] The vector formed by the new interest difference in the branch circuits of the transformer substation is called the new interest difference vector. If the equivalent branch impedance parameters of the transformer substation voltage management device deviate due to a fault, the calculation will result in a non-zero parameter in the new interest difference vector because incorrect impedance parameters are used, leading to errors caused by incorrect branch parameters.

[0141] If the new information difference of a certain branch is not zero, it indicates that the actual electrical quantities (voltage, current, power) of the branch deviate from the model prediction, which is the core basis for judging branch anomalies.

[0142] Suppose a parameter error exists on the nth branch of a radial distribution network. The new interest differences for each branch in the distribution area are calculated, forming a new interest difference vector. The non-zero new interest difference elements in this vector are then sorted according to their branch connections, forming an abnormal path starting from the root node (transformer outlet) of the distribution area. The extent of this path corresponds to the location of the abnormal branch. For example, the abnormal path obtained by sorting the non-zero new interest difference elements according to their branch connections is: , where n is the branch number corresponding to the last non-zero new interest difference element, that is, the branch number to which the abnormal path extends; according to the above formula, it can be seen that the non-zero new interest difference path extends all the way to the beginning of the nth branch, while the new interest difference value at the end of the branch is zero. Therefore, it can be determined that there is a parameter error in the nth branch (such as incorrect resistance and reactance parameters). Thus, it can be concluded that the voltage management device in the corresponding transformer area has malfunctioned.

[0143] By taking maintenance-free measures such as adaptive adjustment of control parameters or restarting of the faulty device, the possibility of self-healing of the faulty device is determined. If it cannot heal itself, the relevant fault information, measures taken, and suggestions for subsequent manual handling are automatically pushed to the relevant operation and maintenance personnel, and the above information is uploaded to the remote master station to generate a defect report.

[0144] The embodiments of this application have the following technical effects:

[0145] 1. By using a real-valued transformer area power flow calculation model, the difficulty of power flow calculation is reduced, making it easier to quickly perform iterative calculations around voltage distribution.

[0146] 2. By forming typical distribution area voltage distribution data sets under various scenarios, estimating photovoltaic power output status, and performing edge computing disturbance analysis based on the sensitivity method, the distribution area voltage distribution can be quickly calculated under the conditions of limited measurement points in the distribution network and photovoltaic power output fluctuations. This assists in the generation of device strategies and reduces the investment in communication and measurement resources required for global optimization of distribution network voltage.

[0147] 3. A fault self-diagnosis method for voltage regulators based on the innovation vector approach is used to quickly locate device faults. The faulty device is then attempted to self-heal through maintenance-free measures such as adaptive adjustment of control parameters or restart. If self-healing fails, a defect report is generated and uploaded to the main station to prevent the device fault from impacting the power grid and improve the grid-friendliness of the device.

[0148] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0149] Example 2:

[0150] This application provides a distribution network transformer area voltage management system, including:

[0151] Data acquisition module: used to collect electrical and non-electrical data of the transformer area in real time, store the collected data, preprocess the data, and upload it to the main station according to preset rules;

[0152] Control strategy generation module: used to perform power flow calculations for transformer substations based on electrical and non-electrical quantity data and to correct the rapid power flow results based on the sensitivity method, thereby obtaining the voltage of each node in the transformer substation; and to generate control commands according to the preset target voltage range to adjust the output voltage of the device.

[0153] Fault self-diagnosis and handling module: It is used to analyze the device's operating status by combining control commands and control execution effects, as well as the device fault self-diagnosis method based on the hologram method; attempt fault self-healing; and push fault-related information and handling suggestions.

[0154] The voltage management system for distribution network areas adopts the method described in Example 1 to achieve voltage management in distribution network areas.

[0155] Example 3:

[0156] This embodiment provides an electronic device, including: a memory and a processor;

[0157] The memory is used to store computer programs;

[0158] The processor is configured to invoke the computer program to execute the method as described in Embodiment 1.

[0159] Example 4:

[0160] This embodiment provides a computer-readable storage medium storing a computer program. When the computer program is run on an electronic device, it causes the electronic device to perform the method described in Embodiment 1.

[0161] Example 5:

[0162] This embodiment provides a computer program product, including a computer program that, when run on an electronic device, causes the electronic device to perform the method described in Embodiment 1.

[0163] The specific implementation of the system, electronic device, computer-readable storage medium, and computer program product provided in this application can be referred to the specific embodiments of the above methods, and will not be repeated here.

[0164] Obviously, those skilled in the art should understand that the various units or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device, or fabricating them separately as individual integrated circuit modules, or fabricating multiple modules or steps into a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0165] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for managing voltage in a distribution network area, characterized in that, Voltage regulation devices installed in series on the low-voltage side of distribution network areas include: Data acquisition: Real-time acquisition of electrical and non-electrical quantity data of the transformer area, storage, preprocessing of the acquired data and uploading to the main station according to preset rules; Control strategy generation: Based on electrical and non-electrical quantity data, power flow calculation of the transformer area is carried out and fast power flow result correction based on sensitivity method is performed to obtain the voltage of each node in the transformer area; control commands are generated according to the preset target voltage range to adjust the output voltage of the device; Fault self-diagnosis and handling: By combining control commands and control execution effects, as well as the device fault self-diagnosis method based on the hologram method, the device operating status is judged; fault self-healing is attempted, and fault-related information and handling suggestions are pushed out.

2. The distribution network transformer area voltage management method according to claim 1, characterized in that, The power flow calculation of the distribution area is based on the distribution area voltage calculation model of a typical scenario including distributed photovoltaic. This model divides the distribution area nodes into PQ nodes, PV nodes and balance nodes, and regards the distribution area voltage management device as the adjustable impedance of the power supply network connected to the distribution area. Wherein, the PQ node refers to a node with a given fixed active power P and reactive power Q, and a constant apparent power S = P + Q. PV nodes refer to nodes given P and U; A balanced node refers to a node given U; The calculation process includes: 1) Initialization: Set the initial voltage values, initial power values, admittance matrix, and other parameters required for calculation at each node in the transformer area; 2) Calculate the initial value of the node current: Calculate according to the formula I(0)=S(0) / U(0), where S(0) is the initial value of the node power and U(0) is the initial value of the node voltage; 3) Iteratively calculate the voltage deviation of the node relative to the slack node: using the formula Iterative calculation of voltage deviation ∆U, where Let I(n) be the inverse of the nodal admittance matrix, and let I(n) be the nodal current in the nth iteration. To balance the node voltage; 4) Restore PQ node voltage: Based on the balanced node voltage, restore the voltage deviation ∆U obtained through iteration to the actual PQ node voltage: ; 5) Determine if the node voltages have converged: Check if the node voltages meet the convergence criteria. If they have converged, then proceed according to... Calculate the power of each node after convergence, and output the voltage and apparent power of each node; if convergence is not achieved, reset the initial value of the PQ node voltage to the latest iteration value, and return to step 2) to continue iterating.

3. The distribution network transformer area voltage management method according to claim 2, characterized in that, If the voltage of the PQ node is found to be out of limit, then the node is converted into a PV node and the voltage of the node is set to be equal to its voltage limit. If the calculated apparent power S=P+Q of the PV node exceeds the limit, then the node is converted into a PQ node and its apparent power S is set to the apparent power limit of the node.

4. The distribution network transformer area voltage management method according to claim 1, characterized in that, The fast power flow result correction based on the sensitivity method includes the following steps: Input the typical operating data set of the transformer substation corresponding to the current scenario. The data set includes the active power, reactive power, and voltage values ​​of the load nodes and the active power, reactive power, and voltage values ​​of the distributed photovoltaic nodes under different seasons, weather conditions, and time periods. The status changes of each monitoring point in the monitoring area ; Determine whether the detected state change originates from the power output fluctuation of the photovoltaic monitoring point. If so, supplement the active and reactive power of the photovoltaic at the non-monitoring point through state estimation; otherwise, proceed directly to the next step. judge If the value is less than the preset threshold α, the correction amount is calculated using the sensitivity method: by establishing the sensitivity relationship between the state variable and the control variable, the parameter correction value that needs to be adjusted due to the state change is quickly calculated; if not, a complete power flow iteration calculation is performed and the typical scenario data set is updated. Output power flow distribution status, which includes the voltage and power of each node.

5. The distribution network transformer area voltage management method according to claim 1, characterized in that, The device fault self-testing method based on the neohologram method includes the following steps: Calculate the innovation vector: Innovation vector = Current measurement vector - Current forecast vector, where the current forecast vector is calculated from the power flow at the previous time, and the current measurement vector is obtained from the actual measurement value at the current time; Using the low-voltage side of the distribution transformer as the root node, each line in the transformer area as a tree branch, and the nodes in the transformer area as connecting branches, a new information vector diagram of the transformer area is constructed. Calculate the cumulative innovation vector of the path: From the root node to the terminal load node, calculate the cumulative innovation vector of the path according to the formula. If the cumulative innovation vector of the path is greater than the preset threshold, it is determined that there is bad data in the system and suspicious branch identification is carried out. Reverse calculation of tree branch information: Perform reverse calculation of tree branch information from the terminal node to the root node, calculate the node load information, branch load information, expected branch load information and information difference respectively, where the information difference is the difference between the current information vector of the branch and the expected branch load information; Fault diagnosis: The new interest difference of the branch in the transformer area is formed into a new interest difference vector. The non-zero elements in the new interest difference vector are sorted according to the branch connection relationship to form an abnormal path starting from the root node of the transformer area. The fault of the transformer area voltage management device in the corresponding branch is determined based on the extension range of the abnormal path.

6. The method for managing voltage in distribution network areas according to claim 1, characterized in that, The method further includes: Remote Interaction: Through a remote interaction module based on the MQTT protocol, the device uploads local information to the master station and receives the collaborative control strategy given by the master station based on the power flow calculation of the distribution network, realizing the collaborative control of multiple devices or the optimization of control parameters in the mode of human intervention; the device connects to the intelligent fusion terminal of the distribution area through wired or wireless communication, and realizes data transmission, device status visualization and real-time local control based on the preset communication protocol.

7. A voltage management system for distribution network areas, characterized in that, include: Data acquisition module: used to collect electrical and non-electrical data of the transformer area in real time, store the collected data, preprocess the data, and upload it to the main station according to preset rules; Control strategy generation module: used to perform power flow calculations for transformer substations based on electrical and non-electrical quantity data and to correct the fast power flow results based on the sensitivity method, so as to obtain the voltage of each node in the transformer substation. The device generates control commands based on a preset target voltage range to adjust the device's output voltage. Fault self-diagnosis and handling module: It is used to analyze the device's operating status by combining control commands and control execution effects, as well as the device fault self-diagnosis method based on the hologram method; attempt fault self-healing; and push fault-related information and handling suggestions. The distribution network area voltage management system adopts the method described in any one of claims 1 to 6 to achieve distribution network area voltage management.

8. An electronic device, characterized in that, include: Memory and processor; The memory is used to store computer programs; The processor is configured to invoke the computer program to perform the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed on an electronic device, causes the electronic device to perform the method as described in any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is run on an electronic device, it causes the electronic device to perform the method as described in any one of claims 1 to 6.

Images