Method and system for testing anti-islanding module

Abstract

The invention relates to a test method and system for an anti-islanding module, and belongs to the technical field of distributed power generation systems. According to the method, a distributed power supply system model is obtained through training based on a topological structure of the distributed power supply system and actually-collected operating parameters of the distributed power supply system, and the model outputs parameters corresponding to power grid fault working conditions or different load change working conditions by adjusting input of the model. Simulating a power grid fault working condition or different load change working conditions by the model, and then obtaining performance indexes of the anti-islanding module under the power grid fault working condition or different load change working conditions, namely obtaining the performance indexes of the anti-islanding module under the actual operation environment of the distributed power supply system; the input of the model is actually acquired operation parameters of the distributed power supply system. According to the method, various actual operation environments of the distributed power supply system applied by the anti-islanding module can be provided, and the method is suitable for testing various anti-islanding modules and is good in applicability.

Description

Technical Field

[0001] The invention relates to a testing method and system for an anti-islanding module, belonging to the technical field of distributed power generation systems. Background Art

[0002] Under the current background of energy transformation, low-voltage distributed photovoltaic power generation systems have been increasingly widely used. Many residential houses, small industrial and commercial sites, etc. have been connected to such photovoltaic systems to realize the use of clean energy. However, when the power grid stops working due to a fault or power outage, if the photovoltaic system cannot detect the power outage in time and stop supplying power to the grid, an island effect will occur, which will not only endanger the lives of grid maintenance personnel, but may also damage power equipment and affect the quality of electricity. Therefore, preventing the occurrence of the island effect, that is, anti-islanding, has become an important technical issue that must be faced when distributed power systems such as photovoltaic power generation, wind power, and hydropower are connected to the grid.

[0003] The anti-islanding device or anti-islanding strategy is set up to deal with the islanding effect. The anti-islanding device, also known as the anti-islanding protection device or anti-islanding module, is a device specially used to detect the connection status between the power grid and the distributed generation system. When the power grid fails or is disconnected, the device can detect and cut off the connection between the distributed generation system and the power grid in time to prevent the occurrence of the islanding effect. The anti-islanding performance and reliability of the anti-islanding protection device are crucial in the distributed generation system.

[0004] However, existing detection methods and testing methods for anti-islanding have many limitations: on the one hand, traditional on-site testing methods often rely on specific test equipment, and the equipment functions are relatively simple, making it difficult to simulate complex and changeable actual working conditions to fully verify the anti-islanding performance; on the other hand, although simulation tests based solely on local computers can build a variety of scenarios, they lack close integration with actual on-site operating data, resulting in possible deviations between test results and the actual situation. In addition, data processing and storage capabilities are limited by local hardware, which is inconvenient for large-scale, long-term test data management.

[0005] At present, the anti-islanding modules of low-voltage distributed photovoltaics are being put into market applications on a large scale, and the demand for product testing is urgent. However, the existing test devices for anti-islanding modules have single functions, and the data model is usually inconsistent with the actual situation; and the current test technology update speed is much slower than the development speed of new technologies; at the same time, there are many product manufacturers, and each uses different technologies; the internal test methods of each manufacturer are quite different from the original metering data. If you want to test the standard original metering data model, you need to send the tested equipment to the standard laboratory, which has a long test cycle and high test costs. As a manufacturer of low-voltage distributed photovoltaic anti-islanding modules, users, and competent testing units, they all hope to find an efficient and fast testing method to improve overall work efficiency and reduce production testing costs.

[0006] With the development of cloud computing technology and Internet of Things technology, it is expected to provide a more efficient, accurate and complex-condition adaptable new method for the anti-islanding test of low-voltage distributed photovoltaics, so as to overcome the deficiencies of the existing technologies and better ensure the safe and stable operation of the photovoltaic system. Summary of the Invention

[0007] The purpose of the present invention is to provide a test method and system for an anti-islanding module, so as to solve the problem of poor applicability in the existing anti-islanding module test.

[0008] To achieve the above purpose, the solution of the present invention includes: A test method for an anti-islanding module of the present invention includes the following steps: Adjust the input of the trained distributed power system model so that the model outputs parameters corresponding to grid fault conditions or different load change conditions, so as to simulate the grid fault conditions or different load change conditions by the model, and further obtain the performance indicators of the anti-islanding module under grid fault conditions or different load change conditions; The model is trained based on the topological structure of the distributed power system and the actually collected operation parameters of the distributed power system; the input of the model is the actually collected operation parameters of the distributed power system; the distributed power system model is of the same system type as the distributed power system.

[0009] Further, the distributed power system model selects a distributed photovoltaic system model, and the distributed power system selects a distributed photovoltaic system.

[0010] Further, the operation parameters are collected by voltage transformers, current transformers and temperature sensors deployed at the inverter output side, grid connection point and load access point.

[0011] Further, the operation parameters adopt preprocessed operation parameters; the preprocessing includes filtering, amplification and analog-to-digital conversion.

[0012] Further, the operation parameters include the voltage, current and frequency at the inverter output side, the voltage, current and frequency at the grid connection point, and the voltage, current and frequency at the load access point.

[0013] Further, the performance indicators include the time to trigger the protection action and the accuracy, false detection rate and missed detection rate of island detection.

[0014] Further, the grid fault conditions include short circuit faults, ground faults and voltage sags.

[0015] Further, the different load change conditions include the conditions of resistive load change, inductive load change, capacitive load change and different power factor change conditions.

[0016] Further, the distributed power system model is deployed in the cloud, and the operating parameters are transmitted to the distributed power system model in the cloud via a secure communication link.

[0017] A test system for an anti-islanding module of the present invention includes a processor, and the processor is used to execute a computer program to implement the steps of the test method for the anti-islanding module as described above.

[0018] Advantages of the present invention: The present invention is a pioneering invention that provides a test method for an anti-islanding module. The method trains a distributed power system model based on the topological structure of the distributed power system and the actually collected operating parameters of the distributed power system. By adjusting the input of the model, the input of the model is the actually collected operating parameters of the distributed power system, so that the model outputs the parameters corresponding to the power grid fault conditions or different load change conditions, in order to realize the simulation of the power grid fault conditions or different load change conditions by the model, and then accurately simulate the actual operating environment of the distributed power system. Then, obtain the performance indicators of the anti-islanding module under the power grid fault conditions or different load change conditions, and obtain the performance indicators of the anti-islanding module under two types of complex conditions, namely the power grid fault conditions and different load change conditions, that is, obtain the performance indicators of the anti-islanding module under the actual operating environment of the distributed power system. Among them, the distributed power system model is of the same system type as the distributed power system. This method can accurately simulate the actual operating environment of the distributed power system to which the anti-islanding module is applied, enable the anti-islanding module to be in various actual operating environments of the distributed power system simulated by the model, comprehensively evaluate the performance of the anti-islanding module, and the test method is applicable to the test of various anti-islanding modules, with good applicability, improved test efficiency and accuracy, and can provide reliable guarantee for the safe and stable operation of the distributed power system, and reduce the potential power safety accidents caused by the failure of the anti-islanding protection. Description of the Drawings

[0019] Figure 1 It is a flow block diagram of on-site sampling and cloud simulation testing of a low-voltage distributed photovoltaic anti-islanding module. Detailed Embodiments

[0020] To solve the problems in the background technology, the present invention provides a test method for an anti-islanding module. Considering that different anti-islanding modules are applied to the operating environments of different distributed power systems, this test method uses the actually collected operating parameters to train a distributed power system model to accurately simulate the actual operating environment of the distributed power system, and then obtains the performance indicators of the anti-islanding module under various conditions. This method is applicable to the test of various anti-islanding modules and has good applicability.

[0021] To make the objectives, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0022] An embodiment of a test method for an anti-islanding module: A test method for an anti-islanding module includes the following steps: Based on the topological structure of the distributed power system and the actually collected operating parameters of the distributed power system, a distributed power system model is trained to accurately simulate each component of the distributed power system and the electrical connection relationship between them, and accurately simulate the operating environment of the distributed power system. The input of the distributed power system model is the actually collected operating parameters of the distributed power system, and the output of the distributed power system model is the parameters corresponding to various operating conditions.

[0023] By adjusting the input of the trained distributed power system model, the model is made to output the parameters corresponding to the grid fault condition or different load change conditions, so as to realize the simulation of the grid fault condition or different load change conditions by the model, and obtain the performance indicators of the anti-islanding module under the grid fault condition or different load change conditions. This test method can effectively evaluate the anti-islanding performance of the anti-islanding module applied to the distributed power system, and ensure that the islanding effect can be detected and avoided in a timely and accurate manner under various operating conditions. Among them, the distributed power system model is of the same system type as the distributed power system.

[0024] Specifically, the distributed power system model and the system of the distributed power system can be selected from a distributed photovoltaic system, a distributed wind power system and a distributed hydropower system. Taking the distributed photovoltaic system as an example, the distributed power system model selects a distributed photovoltaic system model, and the distributed power system selects a distributed photovoltaic system.

[0025] Specifically, the actually collected operating parameters of the distributed power system are collected by voltage transformers, current transformers and temperature sensors deployed at the inverter output side, the grid connection point and the load access point. Of course, other existing collection methods can also be used to collect the operating parameters.

[0026] Specifically, the operating parameters are the preprocessed operating parameters to improve the reliability and accuracy of the model, and thus improve the accuracy and precision of the test; among them, the preprocessing includes filtering, amplification and analog-to-digital conversion. Of course, other processing of the operating parameters can also be carried out according to actual needs. Or if accurate operating parameters with negligible noise are obtained through other means, the operating parameters may not be processed additionally.

[0027] Specifically, the operating parameters include the voltage, current and frequency at the inverter output side, the voltage, current and frequency at the grid connection point, and the voltage, current and frequency at the load access point.

[0028] As other embodiments, the operating parameters include the voltage, current, frequency and temperature on the output side of the inverter, the voltage, current, frequency and temperature at the grid connection point, and the voltage, current, frequency and temperature at the load connection point.

[0029] Specifically, the performance indicators include the time to trigger the protection action and the test results. The test results include the accuracy of islanding detection, the false detection rate and the missed detection rate.

[0030] As other embodiments, the performance indicators include the time to trigger the protection action and the accuracy of islanding detection.

[0031] Specifically, the grid fault conditions include short-circuit faults, grounding faults and voltage sags.

[0032] Specifically, the different load change conditions include the conditions of resistive load change, inductive load change, capacitive load change and different power factor change conditions.

[0033] Specifically, the distributed power system model is deployed in the cloud, and the operating parameters are transmitted to the distributed power system model in the cloud through a secure communication link, combining on-site sampling with the powerful data processing and simulation capabilities in the cloud to provide a more efficient, accurate and complex-condition-adaptive test method.

[0034] Taking the low-voltage distributed photovoltaic system as an example, this solution relates to the fields of photovoltaic power generation technology and power system test technology, and specifically provides a method for on-site sampling and cloud simulation testing of the anti-islanding module of a low-voltage distributed photovoltaic system. This method is based on on-site sampling and uses the cloud to simulate and test the anti-islanding module in the low-voltage distributed photovoltaic system. The specific steps are as follows: 1. This method first accurately collects the operating parameters related to the low-voltage distributed photovoltaic system at the site of the low-voltage distributed photovoltaic system. The operating parameters include various data such as voltage, current, frequency, etc., and uses high-precision sensors and advanced data acquisition equipment to ensure the accuracy and real-time nature of sampling.

[0035] 2. The collected data is transmitted to the cloud through a secure communication link. In the cloud, a low-voltage distributed photovoltaic system model that conforms to the actual situation on-site is constructed using the collected on-site data and an advanced simulation model, and at the same time, the anti-islanding module is embedded in this model.

[0036] 3. Multiple complex conditions are set for this model in the cloud through the collected data. The complex conditions are such as different load changes, grid fault types, etc. Under the set complex conditions, a comprehensive simulation test is carried out on the anti-islanding module, and the performance indicators of the anti-islanding module in different scenarios are obtained.

[0037] The test method of the present invention can solve the problems existing in the existing anti-islanding detection methods, such as detection blind spots, impact on power quality, and insufficient adaptability. Through precise simulation tests, it accurately simulates the actual operating environment of the low-voltage distributed photovoltaic system, effectively and comprehensively evaluates the anti-islanding performance of the anti-islanding module of the low-voltage distributed photovoltaic system, improves the test efficiency and accuracy, ensures that the islanding effect can be detected and avoided in a timely and accurate manner under various working conditions, provides a reliable guarantee for the safe and stable operation of the low-voltage distributed photovoltaic power generation system, and reduces the potential power safety accidents caused by the failure of anti-islanding protection.

[0038] The following combines Figure 1 to elaborate in detail on the specific implementation process of the on-site sampling cloud simulation test method for the low-voltage distributed photovoltaic anti-islanding module: I. Deployment and data collection of the on-site sampling module 1. At the site of a suitable low-voltage distributed photovoltaic system (the on-site of low-voltage distributed photovoltaic power generation conditions), high-precision on-site sampling modules are installed at three key positions: the output side of the photovoltaic inverter, the grid connection point, and the main load access line (the main load access point). Data collection is achieved through the on-site collection module. These sampling modules are equipped with various sensors such as voltage sensors, current sensors, and temperature sensors, and can collect multi-dimensional physical quantity data such as voltage, current, and temperature during the operation of the system in real time.

[0039] Specifically, high-precision voltage transformers and current transformers are installed on the output side of the photovoltaic inverter to focus on collecting the AC voltage and current signals output by the inverter, and a temperature sensor is also equipped to monitor the working temperature of the inverter; high-precision voltage transformers and current transformers are installed at the grid connection point to focus on collecting the voltage and current conditions at the connection point between the grid side and the photovoltaic system; the sampling module on the load access line mainly obtains the relevant electrical quantities and temperature information at the load end.

[0040] 2. The on-site sampling module is built-in with a data preprocessing unit, which is used to perform preprocessing operations on the collected raw data, such as filtering, amplification, and analog-to-digital conversion, remove high-frequency interference components, then amplify the weak electrical signals to an appropriate amplitude range, and convert the analog signals into digital signals through an analog-to-digital converter to ensure the accuracy and stability of the data. The processed data is packaged at a certain time interval and transmitted to the cloud server in real time through an Internet of Things communication module (such as a 4G / 5G wireless communication module or a wired communication module based on Ethernet).

[0041] II. Data transmission, model construction, and parameter configuration 1. The cloud server is set up with dedicated data receiving interfaces, such as opening the corresponding TCP / IP communication ports and configuring the receiving interfaces compatible with multiple Internet of Things communication protocols to ensure the smooth reception of data transmitted from each on-site sampling module. The Internet of Things communication protocols include the HPLC communication protocol, the HRF communication protocol, etc.

[0042] 2. The received data is classified and stored in the cloud. Specifically, a distributed file system is used as the cloud data storage infrastructure, and the received data is redundantly backed up on multiple storage nodes. Multiple copies of each data block are saved on different storage nodes to ensure the security and integrity of the data. The data includes operation scenario data and operation interference data.

[0043] 3. Based on the powerful computing resources of the cloud, a low-voltage distributed photovoltaic system model (analog standard model) is constructed using system simulation software. This model accurately simulates the components of the photovoltaic system and their electrical connection relationships according to the general topology of the low-voltage distributed photovoltaic system and the actual system parameters obtained from on-site sampling data. The system simulation software can also be used to construct a communication success rate model to predict the communication success rate between the on-site and the cloud.

[0044] Specifically, the cloud test platform can be used to build a low-voltage distributed photovoltaic system model according to the actual topology of the low-voltage distributed photovoltaic system in this residential community. Key parameters such as the rated power, open-circuit voltage, short-circuit current of the photovoltaic array, and the rated capacity and output voltage range of the inverter are extracted from the on-site sampling data stored in the cloud database, and the photovoltaic array, inverter, and their electrical connection relationships are accurately configured in the low-voltage distributed photovoltaic system model to truly simulate the actual on-site system.

[0045] 4. According to the actual test requirements, the anti-islanding related parameters are flexibly configured in the low-voltage distributed photovoltaic system model. For example, different anti-islanding detection algorithms and their corresponding threshold parameters are set, such as setting the over / under voltage detection algorithm and the frequency upper and lower limit thresholds of the over / under frequency detection algorithm; to simulate different load characteristic situations and grid fault types. Different load characteristic situations include various scenarios such as resistive load changes, inductive load changes, capacitive load changes, and different power factor changes, and grid fault types include short-circuit faults, three-phase short-circuit faults, single-phase ground faults, ground faults, voltage sags, etc. Different fault positions and durations are set for each fault to construct a rich variety of simulation test scenarios.

[0046] III. Cloud Simulation Test Execution 1. Start the simulation test process in the cloud. Use the on-site sampling data stored in the cloud database as real-time input to drive the low-voltage distributed photovoltaic system model to start running according to the preset simulation test scenarios. In each simulation test scenario, simulate different stages such as normal grid operation, fault occurrence, and fault recovery, and observe the output of the anti-islanding module, including key indicators such as whether the islanding formation is accurately detected, the time to trigger the protection action, and the impact on power quality.

[0047] For example, when simulating the single-phase grounding fault scenario of the power grid, the low-voltage distributed photovoltaic system model, based on the set fault location and fault duration, combines the real-time voltage and current data sampled on-site to simulate the changes in electrical quantities of each part of the photovoltaic system at the moment of fault occurrence and during the subsequent process, and observes whether the anti-islanding module can accurately detect the islanding formation within the specified time and trigger the protection action to stop the inverter from supplying power to the grid.

[0048] 2. Under different simulation test scenarios, for each anti-islanding detection algorithm, repeat the simulation test multiple times, collect the detection results (such as whether the islanding is accurately detected, misjudgment or missed judgment situations, etc.) and performance data such as the time to trigger the protection action in each test, and organize and summarize these data to form a detailed test result data set, which is stored in another database table dedicated to analysis in the cloud.

[0049] IV. Analysis and Feedback of Test Results 1. In the cloud, use data analysis and mining algorithms to deeply analyze the test result data set, compare the advantages and disadvantages of different anti-islanding detection algorithms under various working conditions, find out the detection blind spots of each algorithm and the key factors affecting its performance.

[0050] For example, in the cloud, use the decision tree classification algorithm in machine learning. Take different simulation test scenario conditions (such as load type, fault type, etc.) as input features, and take performance indicators such as the detection accuracy rate and misjudgment rate of each anti-islanding detection algorithm as output labels to analyze the test result data set. By constructing a decision tree model, clearly show the performance advantages and disadvantages of different algorithms under various working conditions, and find out the algorithms that are prone to misjudgment and the corresponding problems in some complex load and specific fault combination situations.

[0051] 2. Based on the analysis results, use a data visualization tool to generate a visual test report and present it to the photovoltaic system operation and maintenance personnel in the community on the web side. In the report, bar charts are used to compare the detection accuracies of different algorithms, and line charts are used to show the change trends of the misjudgment rates of each algorithm under different load changes, etc., which is convenient for the operation and maintenance personnel to intuitively understand the situation. At the same time, based on the analysis conclusion, send optimization suggestions to the control end of the on-site photovoltaic system, and the operation and maintenance personnel perform corresponding optimization operations on the anti-islanding module in actual operation according to the feedback information, so as to improve the anti-islanding performance of the entire low-voltage distributed photovoltaic system.

[0052] Use the on-site real-time collected data as the input of the analog quantity standard model to simulate complex working conditions, different fault types, the actual operation site, or various interference data, so as to realize the pre-evaluation of the working performance of the anti-islanding module as the device under test.

[0053] An embodiment of a test system for an anti-islanding module: A test system for an anti-islanding module includes a processor, and the processor is used to execute a computer program to implement the steps of a test method for an anti-islanding module. Among them, the steps of a test method for an anti-islanding module have been described in detail in an embodiment of a test method for an anti-islanding module, and will not be elaborated here.

Claims

1. A testing method for an anti-islanding module, characterized in that, The steps include the following: Adjust the input of the trained distributed power system model so that the model outputs parameters corresponding to grid fault conditions or different load change conditions, to achieve the simulation of the grid fault conditions or different load change conditions by the model, and further obtain the performance indicators of the anti-islanding module under the grid fault conditions or different load change conditions; The model is trained based on the topological structure of the distributed power system and the actually collected operation parameters of the distributed power system; the input of the model is the actually collected operation parameters of the distributed power system; the distributed power system model is of the same system type as the distributed power system.

2. The test method of the anti-islanding module according to claim 1, wherein The distributed power system model selects a distributed photovoltaic system model, and the distributed power system selects a distributed photovoltaic system.

3. The test method of the anti-islanding module according to claim 1, characterized in that The operation parameters are collected by voltage transformers, current transformers and temperature sensors deployed at the inverter output side, grid connection point and load access point.

4. The test method of the anti-islanding module according to claim 1 or 3, characterized in that The operation parameters adopt the preprocessed operation parameters; the preprocessing includes filtering, amplification and analog-to-digital conversion.

5. The testing method of the anti-islanding module according to claim 1 or 3, characterized in that, The operation parameters include the voltage, current and frequency at the inverter output side, the voltage, current and frequency at the grid connection point, and the voltage, current and frequency at the load access point.

6. The test method of the anti-islanding module according to claim 1, characterized in that, The performance indicators include the time to trigger the protection action and the accuracy, false detection rate and missed detection rate of island detection.

7. The test method of the anti-islanding module according to claim 1, characterized in that, The grid fault conditions include short-circuit faults, grounding faults and voltage sags.

8. The test method for the anti-islanding module according to claim 1, wherein The different load change conditions include the conditions of resistive load change, inductive load change, capacitive load change and different power factor change conditions.

9. The test method for the anti-islanding module according to claim 1, wherein, The distributed power system model is deployed in the cloud, and the operation parameters are transmitted to the distributed power system model in the cloud through a secure communication link.

10. A test system for an anti-islanding module, including a processor, characterized in that, The processor is used to execute a computer program to implement the steps of the test method of the anti-islanding module according to any one of claims 1 to 9.

Images