Automatic testing method for reactive power compensation control device

Abstract

The present invention discloses an automatic testing method for a reactive power compensation control device, comprising the following steps: step S1. establishing a test strategy model; step S2. establishing a test result judgment model; step S3. obtaining a control strategy of an AVC control module and inputting it into the test strategy model to obtain a test strategy; step S4. obtaining reactive power compensation operating parameters of a specified reactive power compensation device according to the test strategy by a test device; receiving grid operating parameters obtained by a grid parameter monitoring module monitoring the grid; step S5. inputting the reactive power compensation operating parameters and grid operating parameters obtained in step S4 into the test result judgment model to obtain a reactive power compensation fault type, and feeding the reactive power compensation fault type back to the AVC control module. The present invention can automatically and quickly detect the usage status of a reactive power compensation control device when using multiple types of reactive power compensation control devices.

Description

Technical Field

[0001] The present invention relates to the technical field of reactive compensation control device testing, and more particularly to an automatic testing method for a reactive compensation control device. Background Art

[0002] Reactive power compensation is a technology used in power supply systems to improve the power factor of the power grid, reduce losses in power transformers and transmission lines, increase power efficiency, and improve the power supply environment. It can be said that reactive power compensation devices are essential high-power devices in power supply systems.

[0003] The reactive power compensation device can be an added capacitor or reactive power compensation device such as SVC. Due to the different usage scenarios of reactive power compensation devices, the compensation scheme is also different. A variety of reactive power compensation circuits are mixed and used. Through the switching switch or circuit selection switch, the control end is connected to the reactive power compensation controller. The reactive power compensation controller determines the compensation capacity based on the power factor and reactive power, and then switches the parallel capacitor for reactive power compensation to meet the reactive power requirements of units, transformers, lines, etc. in the power grid, and ensure the voltage quality of the grid connection point.

[0004] For any type of reactive compensation control device, the actual capacity of the reactive compensation control device is positively correlated with the actual voltage. In order for the reactive compensation control device to reach the rated capacity, the reactive compensation control device must reach the rated voltage. In order to ensure the long-term safe operation of the reactive compensation control device, the rated voltage of the reactive compensation control device will be greater than the grid voltage. When the reactive compensation control device is working on the power supply network, it is necessary to frequently and regularly detect the effect of reactive compensation and detect and judge its own lifespan, which is very time-consuming and labor-intensive. Therefore, there is an urgent need for an automatic testing method for reactive compensation control devices, which can automatically and quickly detect the use status of reactive compensation control devices when using multiple types of reactive compensation control devices, effectively support the voltage level of the grid connection point at the time of fault and fault recovery, reduce the voltage recovery time after the fault, and ensure that the weak network structure restores the voltage level as soon as possible. Summary of the Invention

[0005] The purpose of the present invention is to provide an automatic testing method for a reactive compensation control device, which can automatically and quickly detect the usage status of the reactive compensation control device when using multiple types of reactive compensation control devices, effectively support the voltage level of the grid connection point at the time of fault and fault recovery, and reduce the voltage recovery time after the fault, so as to ensure that the weak network structure recovers the voltage level as soon as possible.

[0006] To achieve the above-mentioned objectives, an automatic testing method for a reactive power compensation control device is provided. The reactive power compensation control device used includes an AVC control module, a testing device, several reactive power compensation devices, and a grid parameter monitoring module connected to the power grid. The testing device is used to obtain reactive power compensation operating parameters of one or more reactive power compensation devices and to receive grid operating parameters obtained by the grid parameter monitoring module when monitoring the power grid. The AVC control module is used to automatically control the reactive power compensation device according to a control strategy determined by itself. The method comprises the following steps:

[0007] Step S1. Establish a test strategy model; the test strategy model is used to convert the control strategy of the AVC control module into a control strategy set and then input and output the test strategy;

[0008] Step S2. Establishing a test result judgment model; the test result judgment model is used to input reactive compensation operating parameters and grid operating parameters, and output reactive compensation fault type; the reactive compensation fault type is the main type of fault in the reactive compensation control device;

[0009] Step S3. Obtain the control strategy of the AVC control module and input it into the test strategy model to obtain the test strategy;

[0010] Step S4. The test device obtains the reactive compensation operating parameters of the specified reactive compensation device according to the test strategy; the grid operating parameters obtained by the grid parameter monitoring module monitoring the grid;

[0011] Step S5: Input the reactive compensation operating parameters and the grid operating parameters obtained in step S4 into the test result judgment model to obtain the reactive compensation fault type and feed the reactive compensation fault type back to the AVC control module.

[0012] In particular, the specific process of outputting the test strategy from the test strategy model in step S1 includes the following steps:

[0013] Step A1. Obtaining the control strategy of the AVC control module and converting it into a control strategy set; the control strategy set is a set of parameter variables for controlling the reactive power compensation device by the AVC control module;

[0014] Step A2. Experts specify test strategies for the control strategy set based on experience; the test strategy is a set of parameter variables for controlling one or more reactive compensation device test devices;

[0015] Step A3. Construct a test strategy database based on the control strategy set and the corresponding test strategy;

[0016] Step A4. Extract and randomly assign test strategies from the test strategy database to form a training set and a test set;

[0017] Step A5. Use the existing control strategy set as input to train the convolutional neural network, and then input the test set from step A4 into the trained convolutional neural network for machine learning. The new eigenvalues obtained can be used to predict the control strategy based on the new eigenvalues to obtain the test strategy.

[0018] In particular, the ratio of the training set to the test set in step A4 is 8:2.

[0019] In particular, the specific process of determining the model output reactive power compensation fault type based on the test result in step S2 includes the following steps:

[0020] Step B1. Obtaining reactive compensation operating parameters of the test device and converting them into a reactive compensation operating parameter set; the reactive compensation operating parameter set is a set of reactive compensation operating parameter variables of one or more reactive compensation devices;

[0021] Step B2. Obtaining the grid operation parameters of the test device and converting them into a grid operation parameter set; the grid operation parameter set is a set of grid operation parameter variables of the grid;

[0022] Step B3. Combining the reactive compensation operating parameter set and the grid operating parameter set into a test database;

[0023] Step B4. Experts specify reactive power compensation fault types for the test database based on their experience;

[0024] Step B5. Construct a test result database based on the test database and the corresponding reactive power compensation fault type;

[0025] Step B6. Extract and randomly assign reactive power compensation fault types from the test result database to form a training set and a test set;

[0026] Step B7. Use the existing test database as input to train the convolutional neural network, then input the test set from step B6 into the trained convolutional neural network for machine learning. The new eigenvalues obtained can be used to predict the test database based on the new eigenvalues to obtain the reactive compensation fault type.

[0027] In particular, the ratio of the training set to the test set in step B6 is 8:2.

[0028] In particular, the reactive power compensation fault types include one or more reactive power compensation device abnormalities, AVC control module abnormalities, and power grid abnormalities.

[0029] Particularly, the reactive power compensation control device further comprises a load measuring device; the load measuring device is used to measure load operating parameters of the load in the power grid and transmit the load operating parameters to the testing device.

[0030] In particular, after step B3, a step is added to obtain the load operating parameters of the load measuring device and convert them into a load operating parameter set, and the load operating parameter set is combined with the reactive compensation operating parameter set and the power grid operating parameter set of the original step B3 into a test database; the load operating parameter set is a collection of load operating parameter variables of the load.

[0031] Beneficial effects of the present invention:

[0032] The present invention establishes a test strategy model and a test result judgment model by utilizing a convolutional neural network. The test strategy is obtained through the test strategy model, and the test device can automatically test one or more specified reactive compensation devices scientifically, accurately, reasonably, and quickly according to the test strategy; the reactive compensation fault type is obtained through the test result judgment model, and the reactive compensation fault type is fed back to the AVC control module to restore the fault in time. The reactive compensation fault type may include one or more reactive compensation device abnormalities, AVC control module abnormalities, and power grid abnormalities. The test result judgment model can determine the cause, which is convenient for staff to quickly troubleshoot the problem. The present invention is also configured to be able to automatically and quickly detect the use status of the reactive compensation control device when using multiple types of reactive compensation control devices, and can effectively support the voltage level of the grid connection point at the time of fault and fault recovery, reduce the voltage recovery time after the fault, and ensure that the weak network structure restores the voltage level as soon as possible. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0034] Figure 1 This is an overall workflow diagram of an embodiment of the present invention;

[0035] Figure 2 This is a specific flow chart of the test strategy model outputting the test strategy according to an embodiment of the present invention.

[0036] Figure 3 This is a specific flow chart of the reactive compensation fault type output by the test result judgment model according to an embodiment of the present invention. DETAILED DESCRIPTION

[0037] The preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby making a clearer and more precise definition of the protection scope of the present invention.

[0038] like Figure 1 As shown, an automatic testing method for a reactive power compensation control device is provided. The reactive power compensation control device used includes an AVC control module, a testing device, several reactive power compensation devices, and a grid parameter monitoring module connected to the grid. The testing device is used to obtain reactive power compensation operating parameters of one or more reactive power compensation devices and to receive grid operating parameters obtained by monitoring the grid by the grid parameter monitoring module. The AVC control module is used to automatically control the reactive power compensation device according to a control strategy determined by itself. The method includes the following steps:

[0039] Step S1: Establish a test strategy model. The test strategy model is used to convert the control strategy of the AVC control module into a control strategy set and then input it, and output the test strategy.

[0040] Step S2: Establish a test result judgment model. The test result judgment model inputs reactive power compensation operating parameters and grid operating parameters and outputs reactive power compensation fault types. Reactive power compensation fault types are the primary types of reactive power compensation control device failures; these types include one or more reactive power compensation device anomalies, AVC control module anomalies, and grid anomalies.

[0041] Step S3. Obtain the control strategy of the AVC control module and input it into the test strategy model to obtain the test strategy;

[0042] Step S4. The test device obtains the reactive compensation operating parameters of the specified reactive compensation device according to the test strategy; receives the grid operating parameters obtained by the grid parameter monitoring module monitoring the grid

[0043] like Figure 2 As shown, the specific process of step S1 testing the strategy model outputting the test strategy includes the following steps:

[0044] Step A1. Obtaining the control strategy of the AVC control module and converting it into a control strategy set; the control strategy set is a set of parameter variables for controlling the reactive power compensation device by the AVC control module;

[0045] Step A2. Experts specify test strategies for the control strategy set based on experience; the test strategy is a set of parameter variables for controlling one or more reactive compensation device test devices;

[0046] Step A3. Construct a test strategy database based on the control strategy set and the corresponding test strategy;

[0047] Step A4. Extract and randomly assign test strategies from the test strategy database to form a training set and a test set; the ratio of the training set to the test set is 8:2.

[0048] Step A5. Use the existing control strategy set as input to train the convolutional neural network, and then input the test set from step A4 into the trained convolutional neural network for machine learning. The new eigenvalues obtained can be used to predict the control strategy based on the new eigenvalues to obtain the test strategy.

[0049] like Figure 3 As shown, the specific process of judging the model output reactive power compensation fault type based on the test result in step S2 includes the following steps:

[0050] Step B1. Obtaining reactive compensation operating parameters of the test device and converting them into a reactive compensation operating parameter set; the reactive compensation operating parameter set is a set of reactive compensation operating parameter variables of one or more reactive compensation devices;

[0051] Step B2. Obtaining the grid operation parameters of the test device and converting them into a grid operation parameter set; the grid operation parameter set is a set of grid operation parameter variables of the grid;

[0052] Step B3. Combining the reactive compensation operating parameter set and the grid operating parameter set into a test database;

[0053] Step B4. Experts specify reactive power compensation fault types for the test database based on their experience;

[0054] Step B5. Construct a test result database based on the test database and the corresponding reactive power compensation fault type;

[0055] Step B6. Extract and randomly assign reactive power compensation fault types from the test result database to form a training set and a test set; the ratio of the training set to the test set is 8:2.

[0056] Step B7. Use the existing test database as input to train the convolutional neural network, then input the test set from step B6 into the trained convolutional neural network for machine learning. The new eigenvalues obtained can be used to predict the test database based on the new eigenvalues to obtain the reactive compensation fault type.

[0057] The reactive power compensation control device also includes a load measurement device. The load measurement device is used to measure load operating parameters of the loads in the power grid and transmit them to the test device. After step B3, a step can be added to convert the load operating parameters obtained from the load measurement device into a load operating parameter set. This load operating parameter set is then combined with the reactive power compensation operating parameter set and the power grid operating parameter set from step B3 to form a test database. The load operating parameter set is a collection of load operating parameter variables.

[0058] Although the embodiments of the present invention are described in conjunction with the accompanying drawings, the patent owner may make various changes or modifications within the scope of the appended claims. As long as they do not exceed the scope of protection described in the claims of the present invention, they should be within the scope of protection of the present invention.

Claims

1. A method for automatically testing a reactive power compensation control device, wherein the reactive power compensation control device comprises an AVC control module, a test device, several reactive power compensation devices, and a grid parameter monitoring module connected to a power grid; the test device is used to obtain reactive power compensation operating parameters of one or more reactive power compensation devices and to receive grid operating parameters obtained by monitoring the power grid by the grid parameter monitoring module; the AVC control module is used to automatically control the reactive power compensation device according to a control strategy determined by itself; and the method is characterized in that: The following steps are involved: Step S1. Establish a test strategy model; the test strategy model is used to convert the control strategy of the AVC control module into a control strategy set and then input and output the test strategy; Step S2. Establishing a test result judgment model; the test result judgment model is used to input reactive compensation operating parameters and grid operating parameters, and output reactive compensation fault type; the reactive compensation fault type is the main type of fault in the reactive compensation control device; Step S3. Obtain the control strategy of the AVC control module and input it into the test strategy model to obtain the test strategy; Step S4. The test device obtains the reactive compensation operating parameters of the specified reactive compensation device according to the test strategy; the grid operating parameters obtained by the grid parameter monitoring module monitoring the grid; Step S5: Input the reactive compensation operating parameters and grid operating parameters obtained in step S4 into the test result judgment model to obtain the reactive compensation fault type and feed the reactive compensation fault type back to the AVC control module.

2. The automatic testing method of a reactive power compensation control device according to claim 1, characterized in that: The specific process of step S1 testing the strategy model outputting the test strategy includes the following steps: Step A1. Obtaining the control strategy of the AVC control module and converting it into a control strategy set; the control strategy set is a set of parameter variables for controlling the reactive power compensation device by the AVC control module; Step A2. Experts specify test strategies for the control strategy set based on experience; the test strategy is a set of parameter variables for controlling one or more reactive compensation device test devices; Step A3. Construct a test strategy database based on the control strategy set and the corresponding test strategy; Step A4. Extract and randomly assign test strategies from the test strategy database to form a training set and a test set; Step A5. Use the existing control strategy set as input to train the convolutional neural network, and then input the test set from step A4 into the trained convolutional neural network for machine learning. The new eigenvalues obtained can be used to predict the control strategy based on the new eigenvalues to obtain the test strategy.

3. The automatic testing method of a reactive power compensation control device according to claim 2, characterized in that: The ratio of the training set to the test set in step A4 is 8:

2.

4. The automatic testing method for a reactive power compensation control device according to claim 1, characterized in that: The specific process of determining the model output reactive power compensation fault type based on the test results in step S2 includes the following steps: Step B1. Obtaining reactive compensation operating parameters of the test device and converting them into a reactive compensation operating parameter set; the reactive compensation operating parameter set is a set of reactive compensation operating parameter variables of one or more reactive compensation devices; Step B2. Obtaining the grid operation parameters of the test device and converting them into a grid operation parameter set; the grid operation parameter set is a set of grid operation parameter variables of the grid; Step B3. Combining the reactive compensation operating parameter set and the grid operating parameter set into a test database; Step B4. Experts specify reactive power compensation fault types for the test database based on their experience; Step B5. Construct a test result database based on the test database and the corresponding reactive power compensation fault type; Step B6. Extract and randomly assign reactive power compensation fault types from the test result database to form a training set and a test set; Step B7. Use the existing test database as input to train the convolutional neural network, then input the test set from step B6 into the trained convolutional neural network for machine learning. The new eigenvalues obtained can be used to predict the test database based on the new eigenvalues to obtain the reactive compensation fault type.

5. The automatic testing method for a reactive power compensation control device according to claim 4, characterized in that: The ratio of the training set to the test set in step B6 is 8:

2.

6. The automatic testing method for a reactive power compensation control device according to claim 1, characterized in that: The reactive power compensation fault types include one or more reactive power compensation device abnormalities, AVC control module abnormalities, and power grid abnormalities.

7. The automatic testing method for a reactive power compensation control device according to claim 4, characterized in that: The reactive power compensation control device further comprises a load measuring device; the load measuring device is used to measure load operating parameters of the load in the power grid and transmit them to the testing device.

8. The automatic testing method for a reactive power compensation control device according to claim 7, characterized in that: After step B3, a step is added to convert the load operating parameters of the load measuring device into a load operating parameter set, and the load operating parameter set is combined with the reactive compensation operating parameter set and the power grid operating parameter set in the original step B3 to form a test database; The load operation parameter set is a set of load operation parameter variables of the load.

Images