Power distribution network overvoltage detection, prediction calculation and alarm method based on event processing

Abstract

The invention relates to the technical field of power system fault detection, in particular to a power distribution network overvoltage detection, prediction calculation and alarm method based on event processing, which comprises the following steps: S1, monitoring the node voltage of a power distribution network in real time, and triggering an event when detecting that the voltage variation delta Vi of a node i exceeds a preset threshold value due to load abrupt change or new energy access quantity abrupt change; s2, performing overvoltage detection and alarm judgment; s3, calling a pre-stored SD coefficient matrix, and directly obtaining the voltage variation of a target node j corresponding to the node i through a one-stop calculation formula; s4, whether overvoltage occurs from the node to all target nodes is detected according to the result, and corresponding alarm is given according to the result; directional propagation calculation of a voltage jump event is realized by pre-generating an SD coefficient matrix, an impedance coupling error is eliminated by combining orthogonal rotation transformation, a two-stage judgment mechanism and a dynamic voltage margin are cooperatively optimized, second-level accurate early warning is realized, and the problem of real-time safety prevention and control of a new energy high-permeability power distribution network is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system fault detection, in particular to a power distribution network overvoltage detection, prediction calculation and alarm method based on event processing. BACKGROUND

[0002] The current power distribution network is undergoing a profound transformation from centralized power supply to distributed energy supply, especially with the large-scale access of new energy such as photovoltaic and wind power, the voltage fluctuation problem is more and more serious, and the frequent overvoltage phenomenon poses a serious threat to equipment insulation, system stability and user power supply safety. The traditional power flow calculation method is mostly based on periodic scanning, relying on full network modeling and iterative calculation, which has high calculation cost and long response time, and cannot adapt to the second-level or even millisecond-level voltage mutation scene. At the same time, in order to improve the calculation speed, the simplified model often ignores the coupling relationship between line resistance and reactance, which is easy to cause misjudgment. In addition, the existing event-driven algorithm mostly focuses on data transmission layer optimization, lacks quantitative modeling of voltage event propagation law, and it is more difficult to realize accurate response prediction along the path in the radial topology structure. Therefore, there is an urgent need for a new overvoltage prediction method that can quickly respond and has sufficient modeling accuracy to meet the operation requirements of the power distribution network under the new situation.

[0003] The key problems in the prior art include: 1. Slow response speed: the traditional full-node power flow calculation method cannot meet the prediction response requirements of the second-level or even millisecond-level voltage mutation scene; 2. Large prediction error: the simplified power flow model fails to effectively consider the composite characteristics of line impedance, especially in the high-penetration area of new energy, there is a risk of false alarm and missed alarm; 3. Unable to analyze directionally: the existing method lacks a modeling mechanism for the propagation path of voltage events, making it difficult to accurately predict downstream nodes in a radial power distribution network; 4. Weak event modeling capability: most event processing systems are limited to alarm triggering and are not closely coupled with the physical mechanism of voltage change, lacking the support of mechanism-based calculation.

[0004] In view of the above problems, the present application provides a power distribution network overvoltage detection, prediction calculation and alarm method based on event processing. SUMMARY

[0005] The purpose of the present application is to provide a power distribution network overvoltage detection, prediction calculation and alarm method based on event processing, which realizes directional propagation calculation of voltage mutation events by pre-generating an SD coefficient matrix, eliminates impedance coupling error by combining orthogonal rotation transformation, and realizes second-level accurate early warning through two-level judgment mechanism and dynamic voltage margin collaborative optimization, solving the real-time safety prevention and control problem of high-penetration power distribution network of new energy.

[0006] To achieve the above technical purposes, achieve the above technical effects, the present application is realized by the following technical solutions: A power distribution network overvoltage detection, prediction calculation and alarm method based on event processing, comprising the following steps: S1: Real-time monitoring of power distribution network node voltage, when detecting that node i is caused by load mutation or new energy access mutation, resulting in voltage variation ΔV i Exceeding the preset threshold value triggers an event; S2: Perform overvoltage detection and alarm judgment: If the voltage of point i rises Exceeding the upper limit margin , it is determined that node i exceeds the voltage upper limit and triggers overvoltage alarm; If the voltage of point i drops Exceeding the lower limit margin , it is determined that node i is lower than the voltage lower limit and triggers under-voltage alarm; S3: Call the pre-stored SD coefficient matrix to directly obtain the voltage variation of the target node j corresponding to node i through a one-stop calculation formula: Wherein SD coefficient between node i and j, the target node is defined as the downstream node of node i in the radial power distribution network topology; S4: According to whether the overvoltage result, the corresponding alarm is carried out.

[0007] Further, the SD coefficient matrix is generated by the following offline processing: S3.1: Based on the power flow calculation result, the voltage values V i , V j and line current I of node i and target node j are obtained; S3.2: Build the equivalent concentrated load model of node i-j, and equivalent the distributed load to the concentrated load and the line parameters to the equivalent impedance Z considering resistance R and reactance L; S3.3: Process the load power through the orthogonal linear rotation transformation matrix T: Effective load, Invalid load, Rotation angle, take θ=ωt, ω is the synchronous angular frequency; S3.4: According to the power value after rotation transformation, the voltage variation relationship is derived by using circuit theory, and finally: It is called the power flow calculation SD coefficient, or SD coefficient for short. That is, when the voltage at node i changes, the voltage change at node j can be calculated in one step according to the above formula.

[0008] Furthermore, the steps S3.2-S3.4 specifically include: is the equivalent concentrated load, Z is the load removed The equivalent impedance of the circuit after Equivalent concentrated load calculation: The current relationship equation is as follows: (1) The equivalent concentrated load from i to j is simulated using the following formula; , (2) in yes The complex conjugate of Comprehensive equivalence The active power and reactive power are calculated using the following formula; , (3) or , (4) in yes The complex conjugate of When the node voltage occur When the current changes, set by The ratio changes, and the current by The proportional change; and set the current and The phase angle does not change; at this time, the comprehensive equivalent Active power flowing through and reactive power The voltage after the change and the changed current; current Calculated according to the following formula: , (5) According to the calculation results of the current flow in the previous period, the following formula is used to calculate , : (6) The payload injected into Z at node i and invalid load The rotation calculation of , where T is the orthogonal linear rotation transformation matrix: (7) Here, is the rotation angle, θ=ωt is the rotation angle, ω is the synchronous angular frequency; In subsequent calculations, the rotated and The value is replaced and ; The voltage drop at the target node j is calculated by rotating the above equation and using Equation 6 to obtain the following equation: (8) (9) According to formula (8), the differential relationship between voltage changes is established; Depend on get In addition, according to formula (9) get , and then obtain Small changes: Take the coefficient (10) The following formula is obtained: After sorting, we get: You can get Bundle Define the SD coefficient: (11) Where k is calculated by formula (10); Finally, you can get (12) Here It is called the SD coefficient for power flow calculation, or SD coefficient for short. That is, when the voltage at node i changes, the voltage change at node j can be calculated in one step according to formula (12).

[0009] Furthermore, the voltage margin is set based on the power flow calculation results of the previous time section: Upper limit margin: wherein is the upper limit of voltage of node i, which is a pre-specified value; is the actual voltage, which is a calculated value; Lower limit margin: wherein is the lower limit of voltage of node i, which is a pre-specified value; is the actual voltage, which is a calculated value; The upper limit and lower limit margin of the voltage of all other target nodes j considered from node i are also calculated as follows: Upper limit margin: wherein is the upper limit of voltage of node j, which is a pre-specified value; is the actual voltage, which is a calculated value; Lower limit margin: wherein is the lower limit of voltage of node j, which is a pre-specified value; is the actual voltage, which is a calculated value.

[0010] Further, the alarm judgment adopts a three-level processing mechanism: First-level check: immediately detect the voltage variation ΔV of node i when the event is triggered i whether it exceeds its voltage margin range; Second-level calculation: only when ΔV i exceeds or , call the SD coefficient matrix to calculate the ΔV of the target node j j , and judge whether it exceeds or . .

[0011] Third-level alarm: according to the calculation result, detect whether the voltage of all target nodes from this node is overvoltage, and according to the result, perform corresponding alarm.

[0012] On the other hand, the present application proposes an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and when the program is executed by the processor, the above-mentioned method is realized.

[0013] On the other hand, the present application proposes a computer readable storage medium, which stores computer instructions for executing the above-mentioned method.

[0014] The beneficial effects of the present application are: The application realizes directional rapid propagation prediction of voltage mutation: through a pre-constructed SD (Sensitivity Distribution) coefficient matrix, the voltage influence relationship between nodes is accurately quantified and saved in the database, so that the voltage change of the target node can be calculated one-stop and quickly after the event occurs. Compared with the traditional power flow iteration method, the present scheme does not need to re-model the whole network, the calculation complexity is reduced from O(n 3 ) to O(n), millisecond-level prediction response can be realized, and the real-time performance and emergency response capability of the system are greatly improved; The application fuses impedance coupling and voltage propagation mechanism to improve modeling accuracy: the application decomposes the effective load and the ineffective load through an orthogonal rotation transformation matrix, avoids the model error caused by ignoring resistance-reactance coupling in power flow analysis, retains the information of the complex impedance structure on the load side, and the prediction model error is smaller; The application has good topological adaptability and path directionality: in the radial distribution network structure, the downstream node direction scanning starting from the fault point can be realized according to the propagation path of the SD coefficient, and the adaptability is strong; a two-level judgment mechanism is introduced to improve the stability and sensitivity of the early warning strategy: a primary trigger mechanism based on the voltage mutation of the main node is designed, and a secondary dynamic judgment on the affected nodes is carried out, and a hierarchical response strategy is constructed; The application has engineering deployability and calculation resource friendliness: the pre-computation + online matrix multiplication structure is adopted, without the need of high-performance hardware platform, the deployment can be realized, and it is suitable for edge computing terminals; support for fusion and expansion with other power grid intelligent dispatching modules: the calculation model of the application is modularized, and it is convenient to use with AVC, DMS, load prediction system, energy storage control, etc.

[0015] In summary, the application has made breakthroughs in response speed, prediction accuracy, system adaptability and engineering deployment compared with the prior art, can effectively improve the real-time overvoltage prevention and control capability of high-penetration new energy distribution network, and has important engineering popularization value and social and economic significance.

[0016] Of course, any product implementing the present application does not necessarily need to achieve all the advantages described above. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0018] Figure 1 It is a schematic diagram of the overall framework of the application; Figure 2The schematic diagram of the object node of the present application; Figure 3 The schematic diagram of the equivalent concentrated load simulation circuit. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the protection scope of the present application.

[0020] Embodiment 1 S1: Real-time monitoring of the voltage of the node of the power distribution network, when detecting that the voltage variation ΔV of the node i caused by the load mutation or the mutation of the access amount of the new energy exceeds the preset threshold value, triggering the event; i S2: Performing overvoltage detection and alarm judgment: If the voltage of the node i rises by more than the upper limit margin , it is determined that the node i exceeds the upper limit of the voltage and triggers the overvoltage alarm; If the voltage of the node i falls by more than the lower limit margin , it is determined that the node i is lower than the lower limit of the voltage and triggers the under-voltage alarm; S3: Calling the pre-stored SD coefficient matrix, and directly obtaining the voltage variation of the target node j corresponding to the node i through a one-stop calculation formula: Wherein is the SD coefficient between the nodes i and j, and the target node is defined as the downstream node of the node i in the radial power distribution network topology; S4: Performing corresponding alarm according to whether the overvoltage result.

[0021] Embodiment 2 This embodiment uses the equivalent concentrated load simulation circuit diagram as shown in Figure 3 to describe the connection between each equivalent concentrated load simulation node i and node j. The characteristic is that the load set after the node i is equivalent to the equivalent concentrated load, and the electrical relationship from i to j is simulated by the integrated equivalent line impedance.

[0022] In the figure is the "equivalent concentrated load" used to simulate the original distributed load, and Z is the equivalent impedance of the circuit after the load (Po+Qo) is removed (assuming as the equivalent impedance).

[0023] ​The voltage phase angle setting at node i is set to 0.

[0024] Equivalent concentrated load calculation Current relationship equation (1) The equivalent concentrated load from i to j is simulated using the following equation.

[0025] , (2) Here, is the conjugate complex of

[0026] Inflow integrated equivalent The active power and reactive power of , (3) or , (4) Here, is the conjugate complex of

[0027] When the node voltage changes, it is assumed that the current changes in proportion to , and the current changes in proportion to . And it is assumed that the phase angles of the currents and do not change. That is, at this time, the active power and the reactive power flowing through the integrated equivalent are the changed voltage and the changed current (the phase does not change, only the absolute value changes) current , which are calculated according to the following equation.

[0028] , (5) The calculation of the integrated equivalent between i and j is calculated according to the power flow calculation results (voltage and current at both ends) of the previous time period, as follows. , .

[0029] (6) Effective load injected into Z at node i and the invalid load of the rotation calculation, where T is the orthogonal linear rotation transformation matrix, (7) Here, is the rotation angle, usually θ = ωt, ω is the synchronous angular frequency.

[0030] In subsequent calculations, the values of the rotated and are used instead of and .

[0031] Voltage drop calculation of the target node j Through the above rotation calculation, and by using formula (6), the following equation is obtained.

[0032] (8) (9) According to formula (8), the difference relationship between the voltage changes is established.

[0033] From , the In addition, from formula (9) , the is obtained, and then the small change amount Take the coefficient (10) to obtain the following formula.

[0034] After rearrangement, we get That is, Define the SD coefficient (11) Where k is calculated by formula (10).

[0035] Finally, we can get (12) Here is called the power flow calculation SD coefficient, simply referred to as the SD coefficient. That is, when the voltage of node i changes, the voltage change of node j can be calculated in one step according to formula (12).​

[0036] The off-line calculation result of the power flow of the previous time period can be stored in a DB, and when an event triggers the need for on-line overvoltage prediction / alarm, it can be called.

[0037] Through the above series of analysis, through the pre-processing calculation of the SD coefficient, the voltage rise and fall of node j caused by the voltage change (regarded as an event) of node i can be calculated by a one-step formula, so that the overvoltage prediction / alarm of the downstream nodes caused by the voltage change of a node can be completed according to the previous steps. One-stop power flow calculation for overvoltage prediction / alarm based on event processing.

[0038] The preferred embodiments of the present application disclosed above are only used to help explain the present application. The preferred embodiments do not describe all the details and limit the present application to the specific embodiments described. Obviously, many modifications and changes can be made according to the content of the present application. The present application selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present application, so that those skilled in the art can well understand and utilize the present application. The present application is limited only by the claims and their full scope and equivalents.

Claims

1. A method for detecting, predicting, calculating and alarming overvoltage in a distribution network based on event processing, characterized in that: The following steps are involved: S1: Real-time monitoring of the voltage of the distribution network nodes. When it is detected that the voltage change ΔV of node i is caused by a sudden change in load or a sudden change in the amount of new energy access, i Trigger an event when a preset threshold is exceeded; S2: Execute overvoltage detection and alarm judgment: If the voltage at point i rises Exceeding the upper limit margin , determines that node i exceeds the voltage upper limit and triggers the overvoltage alarm; If the voltage at point i drops Exceeding the lower limit margin , determines that the node i is lower than the voltage lower limit and triggers the undervoltage alarm; S3: Call the pre-stored SD coefficient matrix and directly obtain the voltage change of the target node j corresponding to the node i through the one-stop calculation formula: in is the SD coefficient between nodes i and j, and the target node is defined as the downstream node of node i in the radial distribution network topology; S4: Issue corresponding alarm according to whether there is overvoltage.

2. The method for detecting, predicting, calculating and alarming overvoltage in a distribution network based on event processing according to claim 1, characterized in that: The SD coefficient matrix is ​​generated by the following offline processing: S3.1: Based on the power flow calculation results, obtain the voltage value V of node i and target node j i , V j and line current I; S3.2: Construct an equivalent concentrated load model for node ij, equating the distributed load to a concentrated load and integrating the line parameters into an equivalent impedance Z that takes both resistance R and reactance L into account; S3.3: Process the load power through the orthogonal linear rotation transformation matrix T: is the effective load, is the invalid load, is the rotation angle, take θ=ωt, ω is the synchronous angular frequency; S3.4: Based on the power value after rotation transformation, use circuit theory to deduce the voltage change relationship and finally obtain: It is called the power flow calculation SD coefficient, or SD coefficient for short. That is, when the voltage at node i changes, the voltage change at node j can be calculated in one step according to the above formula.

3. The method for detecting, predicting, calculating and alarming overvoltage in a distribution network based on event processing according to claim 2, characterized in that: The steps S3.2-S3.4 specifically include: is the equivalent concentrated load, Z is the load removed The equivalent impedance of the circuit after Equivalent concentrated load calculation: The current relationship equation is as follows: （1） The equivalent concentrated load from i to j is simulated using the following formula; ， （2） in yes The complex conjugate of Comprehensive equivalence The active power and reactive power are calculated using the following formula; ， （3） or ， （4） in yes The complex conjugate of When the node voltage occur When the current changes, set by The ratio changes, and the current by The proportional change; and set the current and The phase angle does not change; at this time, the comprehensive equivalent Active power flowing through and reactive power The voltage after the change and the changed current; current Calculated according to the following formula: ， （5） According to the calculation results of the current flow in the previous period, the following formula is used to calculate , : （6） The payload injected into Z at node i and invalid load The rotation calculation of , where T is the orthogonal linear rotation transformation matrix: （7） Here, is the rotation angle, θ=ωt is the rotation angle, ω is the synchronous angular frequency; In subsequent calculations, the rotated and The value is replaced and ; The voltage drop at the target node j is calculated by rotating the above equation and using Equation 6 to obtain the following equation: （8） （9） According to formula (8), the differential relationship between voltage changes is established; Depend on get In addition, according to formula (9) get , and then obtain Small changes: Take the coefficient （10） The following formula is obtained: After sorting, we get: You can get Bundle Define the SD coefficient: （11） Where k is calculated by formula (10); Finally, you can get （12） Here It is called the SD coefficient for power flow calculation, or SD coefficient for short. That is, when the voltage at node i changes, the voltage change at node j can be calculated in one step according to formula (12).

4. The method for detecting, predicting, calculating, and alarming overvoltage in a distribution network based on event processing according to claim 1, wherein: The voltage margin is set based on the power flow calculation results of the previous time section: Upper limit tolerance: in is the upper limit of the voltage allowed at node i, which is a pre-specified value; is the actual voltage, is the calculated value; Lower margin: in is the lower limit of the voltage allowed at node i, which is a pre-specified value; is the actual voltage, is the calculated value; The voltage upper and lower margins of all other target nodes j considered starting from node i are also calculated as follows; Upper limit tolerance: in is the upper limit of the voltage allowed at node j, which is a pre-specified value; is the actual voltage, is the calculated value; Lower margin: in is the lower limit of the voltage allowed at node j, which is a pre-specified value; is the actual voltage, and is the calculated value.

5. The method for detecting, predicting, calculating and alarming overvoltage in a distribution network based on event processing according to claim 1, wherein: The alarm judgment adopts a three-level processing mechanism: First level check: Immediately detect the voltage change ΔV at node i when the event is triggered i Whether it exceeds its voltage margin range; Second level calculation: Only when ΔV i Beyond or When , the SD coefficient matrix is ​​called to calculate the ΔV of the target node j j , and judge whether No Exceed or ; Level 3 alarm: Detect whether there is overvoltage from this node to all target nodes based on the calculation results, and issue corresponding alarms based on the results.

6. An electronic device comprising a memory and a processor, characterized in that: The memory stores a computer program, and when the program is executed by the processor, the method according to any one of claims 1 to 5 is implemented.

7. A computer-readable storage medium, characterized in that: The device stores computer instructions for executing the method according to any one of claims 1 to 5.

Images