Fault recording method, diagnosis method, DSP device and system of photovoltaic grid-connected inverter

Abstract

The invention discloses a fault recording method, a diagnosis method, a DSP device and a system of a photovoltaic grid-connected inverter. The wave recording method comprises the following steps: S1, acquiring a fault wave recording enabling flag bit: if the fault wave recording enabling flag bit is an enabling flag, starting the fault wave recording of the photovoltaic grid-connected inverter and entering the step S2; if the fault recording enabling flag bit is a non-enabling flag, the fault recording enabling flag bit is obtained again until the recording sampling deadline is reached; s2, according to a pre-configured photovoltaic grid-connected inverter fault recording channel and a sampling period, real-time fault data of the photovoltaic grid-connected inverter fault recording channel are written into a nonvolatile memory in a grouped mode, and grouped fault data are obtained; and S3, uploading the grouped fault data according to a data uploading instruction so as to complete the fault recording of the photovoltaic grid-connected inverter. According to the method, the fault of the photovoltaic grid-connected inverter can be comprehensively and accurately recorded, so that accurate positioning and diagnosis of the fault are realized.

Description

Technical Field

[0001] The present invention belongs to the field of photovoltaic technology, and in particular relates to a fault recording method, a diagnosis method, a DSP device and a system for a photovoltaic grid-connected inverter. Background Art

[0002] The fault recording of photovoltaic grid-connected inverters plays a vital role in the operation and maintenance of the system. It not only helps engineers to identify and analyze faults in a timely manner, but also effectively ensures the stability and operating efficiency of the photovoltaic system.

[0003] However, existing fault recording technology has some significant shortcomings. First, existing technology cannot accurately capture the exact moment when the fault occurs, which makes fault analysis more complicated. More importantly, the recording content is relatively small, and the recorded information is often not comprehensive. This is particularly prominent when continuous faults occur in the same time period. Early recorded data may be overwritten by subsequent data, resulting in the loss of key information. This makes it difficult for the photovoltaic inverter to reproduce the specific state of the fault after the fault occurs, making fault location more difficult and imprecise. Summary of the invention

[0004] The technical problem to be solved by the present invention is to solve the above-mentioned deficiencies in the prior art and propose a fault recording method, diagnosis method, DSP device and system for photovoltaic grid-connected inverters. The recording method can comprehensively and accurately record the faults of photovoltaic grid-connected inverters, thereby realizing accurate positioning and diagnosis of the faults.

[0005] In a first aspect, the present invention provides a fault recording method for a photovoltaic grid-connected inverter, the method comprising the following steps:

[0006] Step S1: Get the fault recording enable flag:

[0007] If the fault recording enable flag is an enable flag, the photovoltaic grid-connected inverter fault recording is started and step S2 is entered; if the fault recording enable flag is a disable flag, the fault recording enable flag is re-acquired until the recording sampling cut-off time is reached;

[0008] Step S2: according to the pre-configured photovoltaic grid-connected inverter fault recording channel and sampling period, the real-time fault data of the photovoltaic grid-connected inverter fault recording channel is grouped and written into the non-volatile memory to obtain grouped fault data;

[0009] Step S3: uploading the grouped fault data according to the data uploading instruction to complete the fault recording of the photovoltaic grid-connected inverter.

[0010] Furthermore, the step S2 specifically includes the following steps:

[0011] Presetting recording channel parameters and sampling period; the recording channel is a selected channel that can reflect photovoltaic grid-connected inverter faults, and the sampling period is set according to the type of historical faults or potential faults;

[0012] According to the preset recording channel parameters and sampling period, the non-volatile memory is divided into N different storage areas;

[0013] The data groups are cyclically written into N different storage areas to obtain grouped fault data; each group of fault data includes pre-fault data, fault data and post-fault data; wherein, when all storage areas of the non-volatile memory store data and none of them are uploaded, the fault recording enable flag is set to a disable flag; N is a natural number greater than 1.

[0014] Furthermore, in the step S3, uploading the grouped fault data according to the data upload instruction specifically includes the following steps:

[0015] According to the data upload instruction, the grouped fault data is uploaded in the order of group numbers;

[0016] Among them, after the first group of fault data is uploaded, the non-volatile memory switches to the second group of fault data upload according to the real-time received sending completion flag; after the second group of fault data is uploaded, the non-volatile memory switches to the third group of fault data upload according to the real-time received sending completion flag; ...; after the N-1th group of fault data is uploaded, the non-volatile memory switches to the Nth group of fault data upload according to the real-time received sending completion flag, after the Nth group of fault data is uploaded, the non-volatile memory switches to the first group of fault data upload according to the real-time received sending completion flag; N is a natural number greater than 1;

[0017] After any group of data is uploaded, a sending completion flag is set and written into the non-volatile memory.

[0018] Furthermore, in step S2, the real-time fault data of the fault recording channel of the photovoltaic grid-connected inverter is grouped and written into a non-volatile memory, specifically:

[0019] The real-time fault data of the fault recording channel of the photovoltaic grid-connected inverter are grouped and written into the ferroelectric memory.

[0020] Furthermore, after step S3, the method further includes:

[0021] Step S4: Determine whether all the fault data in the group has been uploaded:

[0022] If the judgment result is that all the fault data of the group have been uploaded, the fault recording enable flag is set to the enable flag, and then steps S2 to S4 are repeated until the recording sampling cut-off time is reached.

[0023] In a second aspect, the present invention provides a fault diagnosis method for a photovoltaic grid-connected inverter, the method comprising the following steps:

[0024] The fault recording method of the photovoltaic grid-connected inverter described in the first aspect is used to obtain fault recording data of the photovoltaic grid-connected inverter;

[0025] Extracting key features from the fault recording data to obtain extracted features; the key features include current, and / or voltage, and / or power, and / or frequency;

[0026] The extracted features are diagnosed based on a machine learning model to implement fault diagnosis of the photovoltaic grid-connected inverter.

[0027] In a third aspect, the present invention provides a DSP device, the device comprising:

[0028] An acquisition unit, used for acquiring a fault recording enable flag;

[0029] A starting unit, connected to the acquisition unit, and configured to start the photovoltaic grid-connected inverter fault recording when the fault recording enable flag bit acquired by the acquisition unit is an enable flag;

[0030] A writing unit connected to the starting unit is used to write the real-time fault data of the photovoltaic grid-connected inverter fault recording channel into a non-volatile memory in groups according to the pre-configured photovoltaic grid-connected inverter fault recording channel and sampling period to obtain grouped fault data;

[0031] The uploading unit is connected to the writing unit and is used to upload the grouped fault data according to the data uploading instruction to complete the fault recording of the photovoltaic grid-connected inverter.

[0032] Furthermore, the writing unit comprises:

[0033] A setting module, used to pre-set recording channel parameters and sampling period; the recording channel is a selected channel that can reflect photovoltaic grid-connected inverter faults, and the sampling period is set according to the type of historical faults or potential faults;

[0034] A division module, connected to the setting module, for dividing the non-volatile memory into N different storage areas according to preset recording channel parameters and sampling periods;

[0035] A writing module, connected to the dividing module, for cyclically writing the data group into N different storage areas to obtain grouped fault data; each group of fault data includes pre-fault data, fault data and post-fault data;

[0036] Wherein, N is a natural number greater than 1;

[0037] A determination module, connected to the writing module, for determining the data status of the non-volatile memory according to the data written by the writing module;

[0038] The first setting module is connected to the determination module and is used to set the fault recording enable flag to a disable flag when the determination module determines that all storage areas of the non-volatile memory store data and none of them are uploaded.

[0039] Furthermore, the upload unit includes:

[0040] A numbering module, numbering the grouped fault data according to group numbers;

[0041] An uploading module, connected to the numbering module, for uploading the grouped fault data in the order of group numbers according to a data uploading instruction;

[0042] Among them, after the first group of fault data is uploaded, the non-volatile memory switches to the second group of fault data upload according to the real-time received sending completion flag; after the second group of fault data is uploaded, the non-volatile memory switches to the third group of fault data upload according to the real-time received sending completion flag; ...; after the N-1th group of fault data is uploaded, the non-volatile memory switches to the Nth group of fault data upload according to the real-time received sending completion flag, after the Nth group of fault data is uploaded, the non-volatile memory switches to the first group of fault data upload according to the real-time received sending completion flag; N is a natural number greater than 1;

[0043] The second setting module is connected to the uploading module and is used to set a sending completion flag and write it into the non-volatile memory after any group of data is uploaded.

[0044] Furthermore, the device also includes:

[0045] A determination unit, used to determine whether all the fault data of the group has been uploaded;

[0046] The setting unit is connected to the determination unit and is used to set the fault recording enable flag as an enable flag when the determination unit determines that all the fault data of the group has been uploaded.

[0047] In a fourth aspect, the present invention provides a fault recording system for a photovoltaic grid-connected inverter, the system comprising:

[0048] The DSP device described in the third aspect is used to realize fault recording of photovoltaic grid-connected inverters;

[0049] A non-volatile memory connected to the DSP device and used to store fault recording data of the photovoltaic grid-connected inverter;

[0050] Wherein, the non-volatile memory is a FRAM device;

[0051] An ARM device, connected to the DSP device, for issuing a fault recording data upload instruction to the DSP device, and receiving the fault recording data sent by the DSP device;

[0052] The FLASH device is connected to the ARM device and is used to store the fault recording data obtained by the ARM device from the DSP device.

[0053] The beneficial effects of the present invention are as follows:

[0054] Beneficial effects of the present invention:

[0055] 1. Data integrity and accuracy: The present invention ensures the integrity and accuracy of the recorded data through real-time data grouping recording, and timely reflects the operating status of the inverter, thereby facilitating rapid response and fault handling.

[0056] 2. Data persistence: Fault data is written to non-volatile memory to ensure that data is not lost during power outages or system restarts, ensuring its long-term availability.

[0057] 3. Flexible configuration: The present invention supports flexibly configurable recording channels and sampling periods, can be adjusted according to different application requirements, and adapt to a variety of working environments.

[0058] 4. Accurate fault location: The present invention can accurately locate faults, help engineers quickly identify the cause of the fault, and significantly improve maintenance efficiency.

[0059] 5. Key data preservation: Through data caching technology, the present invention can preserve key data before and after the failure, combined with high-frequency real-time sampling and event triggering mechanism to ensure that relevant information is fully recorded when a failure occurs.

[0060] 6. Rich recording content: The present invention adopts data grouping technology to effectively increase the richness of the recording content, classify and store it according to time period and fault type, improve the organization of data, and ensure that more relevant information is recorded in each group.

[0061] 7. Sufficient data support: When a fault occurs, the present invention can quickly and comprehensively record more electrical parameters and operating conditions, providing sufficient data support for subsequent fault analysis.

[0062] 8. Data accuracy: The present invention effectively overcomes the shortcomings of the traditional recording method, such as relatively less content and incomplete recorded information, obtains richer and more comprehensive recording content, provides accurate fault time and detailed fault analysis basis, and significantly improves the accuracy and reliability of fault capture. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 Schematic diagram of a fault recording method for a photovoltaic grid-connected inverter in an embodiment of the present invention;

[0064] Figure 2 1 is a framework diagram of a fault recording system for a photovoltaic grid-connected inverter in an embodiment of the present invention;

[0065] Figure 3 A schematic diagram of the interaction of fault recording data of a photovoltaic grid-connected inverter in an embodiment of the present invention;

[0066] Figure 4 A schematic diagram of calculation of a stopping point of power-on fault recording data in an embodiment of the present invention;

[0067] Figure 5 is a data processing logic flow chart of a photovoltaic grid-connected inverter in an embodiment of the present invention;

[0068] Figure 6 A schematic diagram of fault screening logic of a photovoltaic grid-connected inverter in an embodiment of the present invention;

[0069] Figure 7 This is a flow chart of the storage of fault recording data in an embodiment of the present invention;

[0070] Figure 8 A logic diagram for processing the sending group number of the fault recording data not sent before power failure in an embodiment of the present invention;

[0071] Fig. 9 It is a logic flow chart of fault recording and anti-locking in an embodiment of the present invention;

[0072] Fig.10 A schematic diagram of fault recording storage in an embodiment of the present invention;

[0073] Fig.11 is a ferroelectric address allocation diagram in an embodiment of the present invention;

[0074] Fig.12 This is a flow chart of fault recording storage in an embodiment of the present invention;

[0075] Fig.13This is an example diagram of a fault recording in an embodiment of the present invention;

[0076] Fig.14 This is another example diagram of fault recording in an embodiment of the present invention;

[0077] Fig.15 A schematic diagram of different stopping points in another example diagram of a fault recording in an embodiment of the present invention;

[0078] Fig.16 Schematic diagram of switching three groups of fault recording in an embodiment of the present invention;

[0079] Fig.17 This is a processing logic flow chart after data storage is completed in an embodiment of the present invention;

[0080] Fig.18 This is a processing flow chart after data transmission is completed in an embodiment of the present invention;

[0081] Fig.19 Schematic diagram of a fault recording device for a photovoltaic grid-connected inverter in an embodiment of the present invention;

[0082] Among them, the figure marks are: 10, acquisition unit, 20, start unit, 30, writing unit, 40, upload unit. DETAILED DESCRIPTION

[0083] In order to enable those skilled in the art to better understand the technical solution of the present invention, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0084] It should be understood that the specific embodiments and drawings described herein are only used to explain the present invention rather than to limit the present invention.

[0085] It can be understood that, in the absence of conflict, the various embodiments of the present invention and the various features in the embodiments can be combined with each other.

[0086] It can be understood that, for the convenience of description, the drawings of the present invention only show the parts related to the present invention, while the parts irrelevant to the present invention are not shown in the drawings.

[0087] It can be understood that each unit and module involved in the embodiments of the present invention may correspond to only one physical structure, or may be composed of multiple physical structures, or multiple units and modules may be integrated into one physical structure.

[0088] It can be understood that, without conflict, the functions and steps marked in the flowcharts and block diagrams of the present invention may occur in an order different from that marked in the drawings.

[0089] It is understood that the flowcharts and block diagrams of the present invention illustrate the possible architectures, functions, and operations of the systems, devices, equipment, and methods according to the various embodiments of the present invention. Each box in the flowchart or block diagram may represent a unit, module, program segment, or code, which contains executable instructions for implementing the specified functions. Moreover, each box or combination of boxes in the block diagram and flowchart may be implemented by a hardware-based system that implements the specified functions, or may be implemented by a combination of hardware and computer instructions.

[0090] It can be understood that the units and modules involved in the embodiments of the present invention can be implemented by software or hardware. For example, the units and modules can be located in a processor.

[0091] Embodiment 1:

[0092] like Figure 1 As shown, this embodiment provides a fault recording method for a photovoltaic grid-connected inverter. The fault recording method for a photovoltaic grid-connected inverter provided in this embodiment is suitable for real-time monitoring and fault diagnosis of photovoltaic power generation systems, and can timely capture the operating status and fault information of the inverter, helping engineers to quickly locate problems and reduce downtime. In addition, the method can also be used for performance evaluation, maintenance, accident investigation and compliance inspection to ensure the reliability and safety of the system, while improving users' trust and satisfaction with photovoltaic power generation systems.

[0093] The method comprises the following steps:

[0094] Step S1: Get the fault recording enable flag:

[0095] If the fault recording enable flag is an enable flag, the fault recording of the photovoltaic grid-connected inverter is started and step S2 is entered; if the fault recording enable flag is a disable flag, the fault recording enable flag is reacquired until the recording sampling cutoff time is reached. The fault recording enable flag is a control parameter used to enable or disable the fault recording function of the photovoltaic grid-connected inverter. When the flag is set to the "enabled" state, the system will record the changes in key electrical parameters (such as current, voltage and power) when a fault occurs for subsequent analysis and fault diagnosis; and when the flag is set to the "disabled" or "disabled" state, the recording function will stop. By flexibly configuring the flag, users can adjust the activation state of the recording function according to actual needs.

[0096] Step S2: According to the pre-configured photovoltaic grid-connected inverter fault recording channel and sampling period, the real-time fault data of the photovoltaic grid-connected inverter fault recording channel is grouped and written into the non-volatile memory to obtain grouped fault data.

[0097] When writing the real-time fault data of the fault recording channel of the photovoltaic grid-connected inverter into non-volatile memory, the principles of grouping are mainly based on time, fault type and data characteristics. First, the data is grouped according to the sampling period, such as one group per second, per minute or per hour, to ensure that the fault data in the same time period can be stored centrally. Secondly, the fault data is classified according to the fault type (such as overvoltage, overcurrent, short circuit, etc.) to facilitate subsequent analysis and fault diagnosis. The data in each group can also be prioritized according to its importance to ensure that key parameters (such as sudden changes in voltage and current) are recorded first. In addition, for specific fault events, trigger conditions can be set. When the fault data exceeds a certain threshold, a new data group is automatically generated to capture important fault information. In this way, grouping not only improves the storage efficiency of data, but also enhances the pertinence and effectiveness of subsequent analysis.

[0098] The setting principle of the fault recording channel of the photovoltaic grid-connected inverter adopts the recording channel setting in the existing technology, which is mainly based on the comprehensive monitoring requirements of the system operation status, including the selection of key electrical parameters (such as current, voltage, power and temperature) to ensure that the key data when the fault occurs can be captured in time. In addition, the configuration of the channel also takes into account the characteristics of different fault modes, the working environment of the system and the specific requirements of the application scenario to achieve efficient data collection and fault diagnosis, thereby improving the reliability and maintenance efficiency of the system. The recording channel is set in the control unit inside the photovoltaic grid-connected inverter to directly monitor key electrical parameters (such as current, voltage and power), and divides the non-volatile memory into N different storage areas according to the pre-set recording channel parameters and sampling period, so as to achieve effective storage and management of different fault data.

[0099] In step S2, the real-time fault data of the fault recording channel of the photovoltaic grid-connected inverter is grouped and written into the non-volatile memory, specifically:

[0100] The real-time fault data of the fault recording channel of the photovoltaic grid-connected inverter are grouped and written into the ferroelectric memory.

[0101] The step S2 specifically includes the following steps:

[0102] Presetting recording channel parameters and sampling period; the recording channel is a selected channel that can reflect photovoltaic grid-connected inverter faults, and the sampling period is set according to the type of historical faults or potential faults;

[0103] According to the preset recording channel parameters and sampling period, the non-volatile memory is divided into N different storage areas;

[0104] The data groups are cyclically written into N different storage areas to obtain grouped fault data; each group of fault data includes pre-fault data, fault data and post-fault data; wherein, when all storage areas of the non-volatile memory store data and none of them are uploaded, the fault recording enable flag is set to a disable flag;

[0105] N is a natural number greater than 1.

[0106] Step S3: uploading the grouped fault data according to the data uploading instruction to complete the fault recording of the photovoltaic grid-connected inverter.

[0107] In step S3, uploading the grouped fault data according to the data upload instruction specifically includes the following steps:

[0108] According to the data upload instruction, the grouped fault data is uploaded in the order of group numbers;

[0109] Among them, after the first group of fault data is uploaded, the non-volatile memory switches to the second group of fault data upload according to the real-time received sending completion flag; after the second group of fault data is uploaded, the non-volatile memory switches to the third group of fault data upload according to the real-time received sending completion flag; ...; after the N-1th group of fault data is uploaded, the non-volatile memory switches to the Nth group of fault data upload according to the real-time received sending completion flag, after the Nth group of fault data is uploaded, the non-volatile memory switches to the first group of fault data upload according to the real-time received sending completion flag; N is a natural number greater than 1;

[0110] After any group of data is uploaded, a sending completion flag is set and written into the non-volatile memory.

[0111] Step S4: Determine whether all the fault data in the group has been uploaded:

[0112] If the judgment result is that all the fault data of the group have been uploaded, the fault recording enable flag is set to the enable flag, and then steps S2 to S4 are repeated until the recording sampling cut-off time is reached.

[0113] Existing recording methods often record relatively little content, and the recorded information is not comprehensive enough. However, this embodiment effectively increases the richness of the recorded content by adopting data grouping technology and non-volatile memory. Specifically, data grouping technology can classify and store real-time fault data according to time periods and fault types, which not only improves the organization of the data, but also ensures that more relevant information can be recorded in each group. In addition, the fault data is written to the non-volatile memory to ensure that the data will not be lost when the power is off or the system is restarted, thereby ensuring long-term availability and integrity. This combination enables the system to quickly and comprehensively record more electrical parameters and operating conditions when a fault occurs, providing more sufficient data support for subsequent fault analysis.

[0114] This embodiment includes three steps:

[0115] Step 1: Configure the channels and sampling period for fault recording of the photovoltaic grid-connected inverter. The fault recording software records the faults of each channel and stores them in the ferroelectric storage chip. Based on the fault recording enable flag, the stored data is cyclically recorded and the storage end flag is updated.

[0116] Step 1 is as follows:

[0117] A) According to the ferroelectric memory (storage space 1MBit), the fault recording space is divided. A set of fault recording data storage includes: the recording content includes the channel ID (2 bytes), the time when the fault occurs (8 bytes), the channel content (598*2=1918 bytes) and the fault recording sending completion flag (2 bytes), a total of 1930 bytes. Since the ferroelectric storage content stores other data in addition to the fault recording, it can store up to 3 sets of fault recording data. According to the fault recording write group number, the data write address is calculated.

[0118] B) When the fault recording enable flag is valid, the fault recording software first writes the ID of the configured channel into the ferroelectric memory, and then writes the real-time data of the channel into the ferroelectric memory at a certain period (sampling period * 62.5). When a fault occurs and the fault attribute is recording, the fault recording stop flag is triggered, and the recording stops after continuing to record 319 data points.

[0119] C) When recording of one group is completed, the writing group number is switched (after the first group of records is completed, it switches to recording the second group, after the second group of records is completed, it switches to the third group, after the third group of records is completed, it switches to the first group).

[0120] D) After the first set of records is completed, the second set of records is performed. If a new fault attribute of a recording fault is detected within the time of △T, the second set of records is continued. Otherwise, the fault recording enable flag is set to disabled and the first set of fault recording end flag is updated. The logic of the second and third sets is the same.

[0121] E) When continuous faults occur, if all three groups have been recorded but have not been sent, the fault recording enable flag is set to disabled and no fault recording is performed.

[0122] F) When the PV grid-connected inverter is in grid-connected state, and any set of data recording is completed, and no data upload instruction is received, and this situation lasts for more than or equal to 20s, the fault recording enable flag is set to enable and recording is started.

[0123] Step 2: The fault recording software reads the corresponding data packet from the ferroelectric memory based on the received data upload instruction and uploads it.

[0124] The step 2 is specifically as follows:

[0125] A) When a data upload instruction is received, the first address and the first address of the timestamp are read according to the sending group number allocation, and all fault recording data of the channel number are uploaded according to the channel number of the upload instruction.

[0126] B) When a complete set of fault recordings is uploaded, set the sending group number switch (after the first group is sent, switch to sending the second group, after the second group is sent, switch to the third group, after the third group is sent, switch to the first group), set the sending completion flag, and write it into the iron electric fixed address.

[0127] Step 3: After a complete set of fault recording data is uploaded to the communication board, set the fault recording enable flag to enable.

[0128] The following is an explanation with a specific embodiment:

[0129] 1. Power-on fault recording data processing:

[0130] The processing logic of power-on fault recording data is as follows: Figure 5 As shown. In the initialization processing stage, the fault recording channel and sampling period will be set to the default value, and the group number recorded and sent will be uniformly set to the first group. Then, the system will determine whether it is necessary to send the fault recording data before power failure for processing. The specific processing process is as follows:

[0131] 1) Read the first group of fault recording transmission completion flag stored in the ferroelectric fixed address. If the flag is 0, the power-on fault recording transmission count is +1, and the stop point of this group of data is calculated, and the first group of fault recording full write flag is set to 0xFF.

[0132] 2) Stop point calculation: judge the 12th bit of fault status word 2, and the last point that is 1 is the stop point. Figure 4 shown.

[0133] 3) The judgment logic of the second and third groups is the same as steps 1) and 2).

[0134] 4) If the power-on fault recording sending count is 0, set the software storage end flag 1 and software storage end flag 2 to 0, and the fault recording storage enable flag to 1, otherwise, set the fault recording storage enable flag to 0. The power-on fault recording sending count is 0, which means that before the last power failure, the fault recording data has been transmitted to the communication board. If no data is sent during this power-on, the power-on fault recording sending count is 0, and no data needs to be sent. The fault recording is directly enabled to record data.

[0135] 2. Fault screening

[0136] Fault screening logic such as Figure 6 As shown, the specific process is as follows:

[0137] 1) When the fault recording enable flag is 1 and the fault recording stop has been triggered flag is 0, fault screening is performed, otherwise no screening is performed;

[0138] 2) When the current value of module status word 20 to 29 is not equal to the backup value, the 16 bits of the module status word are judged in turn:

[0139] 3) Determine whether the current value of each bit of the module status word is 1 and the backup value is 0;

[0140] 4) Whether the fault attribute corresponding to the bit is an attribute that can trigger fault recording;

[0141] 5) If both conditions in 2) are met at the same time, the fault recording trigger flag is set to 1 and the loop is exited;

[0142] 6) Back up the module status word.

[0143] Among them, fault attributes are divided into four types: no waveform recording when alarm is triggered, no waveform recording when shutdown is triggered, waveform recording when alarm is triggered, and waveform recording when shutdown is triggered.

[0144] 3. Fault recording storage and processing

[0145] Fault recording storage processing includes 4 parts, such as Figure 7As shown. Data not sent before power failure is sent, communication failure occurs (no fault recording reading instruction is received after recording is completed), anti-locking processing, fault recording data storage, and processing after storage is completed. In the process of processing data not sent before power failure, it is first necessary to identify and respond to possible communication failures. This failure usually occurs when the system fails to receive the fault recording reading instruction after recording is completed. In order to prevent the system from being locked due to failure to communicate normally, anti-locking processing must be implemented. This measure ensures that the system can remain in operation and continue to perform necessary operations. Next, the fault recording data will be safely stored for use in subsequent analysis. After the storage process is completed, the system will perform subsequent processing, including data verification and collation, to ensure the integrity and accuracy of the data, thereby providing reliable information support for subsequent fault analysis and decision-making. This series of processing steps is designed to minimize the risk of data loss due to communication failures and ensure the stability and reliability of the system.

[0146] The specific process of fault recording storage processing is as follows:

[0147] 1) Fault recording data group number selection processing (main interrupt) not sent before power failure:

[0148] like Figure 8 As shown, when the power-on fault recording sending count is not 0, the fault data before power failure is sent.

[0149] If the full mark of the first group is 0xFF, the sending group number is set to the first group; otherwise, if the full mark of the second group is 0xFF, the sending group number is set to the second group; otherwise, if the full mark of the third group is 0xFF, the sending group number is set to the third group. The sending group number processing logic is shown in the figure.

[0150] 2) In case of communication failure, the fault recording system will prevent locking:

[0151] like Fig. 9 As shown, when connected to the grid, when the fault recording storage end flag is not all 1, and the communication instruction is not a fault recording reading instruction, counting starts. When the count value is greater than 20s, the event record records this communication fault event, and sets the fault recording enable flag to 1, the count value is cleared, the three groups of full flags are set to 0, and the data transfer flag is set to 1.

[0152] 3) Fault recording data storage

[0153] DSP sends the recorded content to the ferroelectric memory via SPI. The storage content, storage timing and storage format are as follows Fig.10 As shown in Figure 1. The ferroelectric address allocation is as follows Fig.11 shown.

[0154] like Fig.12As shown, the fault recording storage process includes: recording the variable index or variable content according to the number of fault recording points; when a recording fault occurs, the recording timestamp and the sending completion flag are cleared and the stop point is calculated; after recording to the stop point, the recording completion flag is set and the recording group number is switched.

[0155] When the fault recording count is equal to the fault recording sampling period, and the fault recording enable flag is enabled, select the corresponding record first address and timestamp first address according to the record group number, and then perform the following processing:

[0156] A) Data writing into ferroelectric: When the number of data stored in the fault recorder is 0, the variable index is stored; when the number of data stored in the fault recorder is 1 to 959, the variable content is stored and a circular recording is performed.

[0157] B) Fault occurrence processing: When a fault occurs, record the timestamp, record the sending completion flag (0), and calculate the stop point.

[0158] B.1) When a fault occurs, two situations are distinguished based on the number of recorded points:

[0159] Case 1: If the fault occurs at the beginning of data recording (the recording point is never greater than or equal to 640), the stop update point = 959.

[0160] Specific as Fig.13 As shown, it stops at position 959.

[0161] Case 2: The calculation formula of the stop point is as follows:

[0162] The first way, such as Fig.14 As shown, the stop update point = storage array index + 1 + 319.

[0163] That is, a set of fault recordings records a total of 960 data points, and the data set consists of 1 variable subscript + 959 variable contents, and the 959 variable contents are divided into 2 / 3 recorded before the fault occurs and 1 / 3 recorded after the fault, so recording stops after recording 319 points.

[0164] The second method, such as Fig.15 As shown, the stop update point = 320-(960-1-storage array index).

[0165] C) Processing of the last point recorded: When recording reaches the stop point, set the full mark of this group and the group number of the next group to be written, and set the number of fault recording storage data to 0. Fig.16After the first group of fault data is uploaded, the non-volatile memory switches to the second group of fault data upload according to the real-time received send completion flag; after the second group of fault data is uploaded, the non-volatile memory switches to the third group of fault data upload according to the real-time received send completion flag; after the third group of fault data is uploaded, the non-volatile memory switches to the first group of fault data upload according to the real-time received send completion flag.

[0166] 4) Processing after data storage:

[0167] When any of the three groups is full, counting starts. When the count value reaches △t (△t=62.5*sampling period*960*2 / 3, value range [40,180]ms), no new recording fault is detected, and the fault recording storage enable flag is set to disabled. According to the sending group, the fault recording storage flag 1 and the fault recording storage flag 2 are set. If a new fault is detected within △t, the current group will be recorded. And so on. Fig.17 As shown, the processing after data storage is completed.

[0168] 4. Upload fault recording data

[0169] When the communication board detects that the fault recording storage flag bit 1 and the fault recording storage flag bit 2 have changed, it sends a fault recording read instruction to the DSP;

[0170]

[0171] When the DSP receives the read command, it selects the fault recording first address and timestamp first address according to the read command and the fault recording sending group;

[0172] When reading data from the ferroelectric and uploading it to the communication board, the timestamp + all 960 data of a variable are transmitted in one frame. First read the timestamp, then use a for loop to read all the data of a variable 960 times before starting to send it. The format of the sent data is shown in the table below.

[0173]

[0174]

[0175] When the data transmission of 12 channels is completed, the fault recording storage enable flag is set to enable, and the transmission group is changed, the transmission completion flag of the group is set to 0xFF, and it is written into the ferroelectric fixed address. The processing flow after the data transmission is completed is as follows: Fig.18 shown.

[0176] A preferred embodiment of a photovoltaic grid-connected inverter fault recording method is provided below, comprising the following steps:

[0177] (1) First, the host computer configures the channel and sampling period of the photovoltaic grid-connected inverter fault recording, and the DSP fault recording software performs real-time cyclic recording on each of the channels (for example, the selected channels are 0, 2, 4, 8, 10, 12, 14, 16, 18, 20, 91, 92) according to the sampling period (for example, if the sampling period setting value is 1, the sampling interval is 62.5us);

[0178] (2) When a fault occurs whose attribute is alarm recording or shutdown recording (for example, the attribute of bus short circuit fault is shutdown recording), the recording stop point is calculated based on the fault recording stop signal and the current number of recording points (for example, the current number of points is 720), that is, the data points from 720 to 959 are recorded first, and then the data points are recorded from 1, for a total of 319 data points).

[0179] (3) The fault wave recording continues. When the current recording point number is equal to the stop point, the recording stops, indicating that all the data for this fault (bus short circuit fault) has been written into the ferroelectric memory.

[0180] (4) After a complete set of fault recording data is written into the ferroelectric:

[0181] (4.1) If no new recording fault occurs within the time of △T, DSP will transmit the fault recording completion information to ARM.

[0182] (4.2) If no new recording fault occurs within the time of △T (such as 12V power failure of the main control board), the DSP will continue to write data to the next group of fault recording storage space of the iron electric. After writing this group, it will pass the completion information of the previous group of fault recording to ARM.

[0183] (5) After ARM receives the recording completion information transmitted by DSP, it asks DSP for the recording data of the last fault (bus short circuit).

[0184] (6) When DSP receives the read instruction from ARM, it first reads the fault recording data (bus short circuit fault) from the ferroelectric memory and then sends it to ARM. After receiving the data, ARM transfers it to Flash.

[0185] (7) After the first set of data is transmitted to ARM, DSP will transmit the second set of (main control board 12V power failure) storage end information to ARM. After receiving the information, ARM will issue a read instruction, and DSP will transmit the (main control board 12V power failure) fault data to ARM. Since the time for each set of data to be transmitted to the communication board is about 3s, but the main control board 12V power failure cannot last for 3s, the DSP will be completely powered off. Therefore, the data cannot be fully transmitted to ARM before the power failure. Therefore, when the power is on next time, DSP detects that there is fault recording data before the last power failure that has not been uploaded to ARM, and will upload the data again.

[0186] (8) When a person receives the fault recording instruction from the host computer, ARM reads the data from the Flash and sends it to the host computer.

[0187] The sampling rate of this embodiment can be selected as [1,520], and the data can be stored at the fastest speed of 16KHz, which greatly shortens the storage interval of each recorded data, that is, greatly improves the data resolution, makes it easier to reproduce the real-time status of the energy storage converter before and after the fault, and thus greatly improves the accuracy of grid-connected inverter fault location.

[0188] In the selection process of non-volatile memory, in addition to ferroelectric memory (FeRAM), there are other types of memory to consider, such as Flash Memory, Phase Change Memory (PCM) and Magnetic Random Access Memory (MRAM). However, the ferroelectric memory was finally selected in this embodiment because of its unique characteristics. Ferroelectric memory is not only non-volatile and can retain data after power failure, but also has the advantages of fast read and write speed and low power consumption. In addition, the high durability of FeRAM enables it to withstand more read and write cycles, which is suitable for application scenarios where data is frequently updated. Its simple manufacturing process also makes ferroelectric memory compatible with existing CMOS technology, making it easy to integrate into various devices. These characteristics make ferroelectric memory an ideal choice in many applications, especially when fast, reliable and high-performance storage solutions are required.

[0189] The ferroelectric memory in this embodiment stores 3 groups of data, and the Flash stores 30 groups of data, which can ensure that multiple fault data can be saved, and can ensure that the data before the inverter is powered off can be stored in the ferroelectric memory. After the next power-on, the fault data before the power off is uploaded and stored in the Flash, which can also be stored. This embodiment can comprehensively and accurately record the faults of the photovoltaic grid-connected inverter, thereby realizing the precise positioning of the faults.

[0190] Embodiment 2:

[0191] This embodiment provides a fault diagnosis method for a photovoltaic grid-connected inverter, the method comprising the following steps:

[0192] The fault recording method of the photovoltaic grid-connected inverter described in Example 1 is used to obtain fault recording data of the photovoltaic grid-connected inverter;

[0193] Extracting key features from the fault recording data to obtain extracted features; the key features include current, and / or voltage, and / or power, and / or frequency;

[0194] The extracted features are diagnosed based on a machine learning model to implement fault diagnosis of the photovoltaic grid-connected inverter.

[0195] Embodiment 3:

[0196] like Figure 2 and Fig.19 As shown, this embodiment provides a DSP (the full name of DSP is Digital Signal Processor) device, and the device includes:

[0197] An acquisition unit 10 is used to acquire a fault recording enable flag;

[0198] A starting unit 20, connected to the acquisition unit 10, is used to start the photovoltaic grid-connected inverter fault recording when the fault recording enable flag bit acquired by the acquisition unit is an enable flag;

[0199] The writing unit 30 is connected to the starting unit 20 and is used to write the real-time fault data of the fault recording channel of the photovoltaic grid-connected inverter into the non-volatile memory in groups according to the pre-configured photovoltaic grid-connected inverter fault recording channel and sampling period to obtain grouped fault data;

[0200] The uploading unit 40 is connected to the writing unit 30 and is used to upload the grouped fault data according to the data uploading instruction to complete the fault recording of the photovoltaic grid-connected inverter.

[0201] As a specific implementation, the writing unit 30 includes:

[0202] A setting module, used to pre-set recording channel parameters and sampling period; the recording channel is a selected channel that can reflect photovoltaic grid-connected inverter faults, and the sampling period is set according to the type of historical faults or potential faults;

[0203] A division module, connected to the setting module, for dividing the non-volatile memory into N different storage areas according to preset recording channel parameters and sampling periods;

[0204] A writing module, connected to the dividing module, for cyclically writing the data group into N different storage areas to obtain grouped fault data; each group of fault data includes pre-fault data, fault data and post-fault data;

[0205] Wherein, N is a natural number greater than 1;

[0206] A determination module, connected to the writing module, for determining the data status of the non-volatile memory according to the data written by the writing module;

[0207] The first setting module is connected to the determination module and is used to set the fault recording enable flag to a disable flag when the determination module determines that all storage areas of the non-volatile memory store data and none of them are uploaded.

[0208] As a specific implementation, the uploading unit 40 includes:

[0209] A numbering module, numbering the grouped fault data according to group numbers;

[0210] An uploading module, connected to the numbering module, for uploading the grouped fault data in the order of group numbers according to a data uploading instruction;

[0211] Among them, after the first group of fault data is uploaded, the non-volatile memory switches to the second group of fault data upload according to the real-time received sending completion flag; after the second group of fault data is uploaded, the non-volatile memory switches to the third group of fault data upload according to the real-time received sending completion flag; ...; after the N-1th group of fault data is uploaded, the non-volatile memory switches to the Nth group of fault data upload according to the real-time received sending completion flag, after the Nth group of fault data is uploaded, the non-volatile memory switches to the first group of fault data upload according to the real-time received sending completion flag; N is a natural number greater than 1;

[0212] The second setting module is connected to the uploading module and is used to set a sending completion flag and write it into the non-volatile memory after any group of data is uploaded.

[0213] As a specific implementation, the DSP device further includes:

[0214] A determination unit, used to determine whether all the fault data of the group has been uploaded;

[0215] The setting unit is connected to the determination unit and is used to set the fault recording enable flag as an enable flag when the determination unit determines that all the fault data of the group has been uploaded.

[0216] Embodiment 4:

[0217] like Figure 2 As shown, this embodiment provides a fault recording system for a photovoltaic grid-connected inverter, the system comprising:

[0218] The DSP device described in Example 3 is used to realize fault recording of photovoltaic grid-connected inverter;

[0219] A FRAM (Ferroelectric Random Access Memory) device, connected to the DSP device, for storing fault recording data of the photovoltaic grid-connected inverter; the FRAM device is a type of non-volatile memory;

[0220] An ARM device, connected to the DSP device, for issuing a fault recording data upload instruction to the DSP device, and receiving the fault recording data sent by the DSP device;

[0221] The FLASH device is connected to the ARM device and is used to store the fault recording data obtained by the ARM device from the DSP device.

[0222] In addition to FRAM (Ferroelectric Random Access Memory), other common non-volatile memories include EEPROM (Electrically Erasable Programmable Read-Only Memory), Flash memory, MRAM (Magnetic Random Access Memory) and ReRAM (Resistance Random Access Memory). FRAM is chosen because of its fast read and write speed, high endurance (able to withstand billions of write and erase cycles), low power consumption (especially in standby mode), good data retention (data can still be saved after power failure), simple programming process (no complicated erase steps required) and excellent environmental adaptability, making it an ideal non-volatile storage solution in many applications.

[0223] The fault recording system in this embodiment includes a main control board and a communication board. The main control board includes a DSP processor and a FRAM storage chip, which is responsible for data collection and preliminary storage. The communication board includes an ARM processor and a FLASH memory, which is responsible for data transfer, management and final storage.

[0224] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present invention, but the present invention is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A fault recording method for a photovoltaic grid-connected inverter, characterized in that: The method comprises the following steps: Step S1: Get the fault recording enable flag: If the fault recording enable flag is an enable flag, the photovoltaic grid-connected inverter fault recording is started and step S2 is entered; if the fault recording enable flag is a disable flag, the fault recording enable flag is re-acquired until the recording sampling cut-off time is reached; Step S2: according to the pre-configured photovoltaic grid-connected inverter fault recording channel and sampling period, the real-time fault data of the photovoltaic grid-connected inverter fault recording channel is grouped and written into the non-volatile memory to obtain grouped fault data; Step S3: uploading the grouped fault data according to the data uploading instruction to complete the fault recording of the photovoltaic grid-connected inverter.

2. The fault recording method of photovoltaic grid-connected inverter according to claim 1, characterized in that: The step S2 specifically includes the following steps: Presetting recording channel parameters and sampling period; the recording channel is a selected channel that can reflect photovoltaic grid-connected inverter faults, and the sampling period is set according to the type of historical faults or potential faults; According to the preset recording channel parameters and sampling period, the non-volatile memory is divided into N different storage areas; The data groups are cyclically written into N different storage areas to obtain grouped fault data; each group of fault data includes pre-fault data, fault data and post-fault data; wherein, when all storage areas of the non-volatile memory store data and none of them are uploaded, the fault recording enable flag is set to a disable flag; N is a natural number greater than 1.

3. The fault recording method of photovoltaic grid-connected inverter according to claim 1, characterized in that: In step S3, uploading the grouped fault data according to the data upload instruction specifically includes the following steps: According to the data upload instruction, the grouped fault data is uploaded in the order of group numbers; Among them, after the first group of fault data is uploaded, the non-volatile memory switches to the second group of fault data upload according to the real-time received sending completion flag; after the second group of fault data is uploaded, the non-volatile memory switches to the third group of fault data upload according to the real-time received sending completion flag; ...; after the N-1th group of fault data is uploaded, the non-volatile memory switches to the Nth group of fault data upload according to the real-time received sending completion flag, after the Nth group of fault data is uploaded, the non-volatile memory switches to the first group of fault data upload according to the real-time received sending completion flag; N is a natural number greater than 1; After any group of data is uploaded, a sending completion flag is set and written into the non-volatile memory.

4. The fault recording method of photovoltaic grid-connected inverter according to claim 1, characterized in that: In the step S2, the real-time fault data of the fault recording channel of the photovoltaic grid-connected inverter is grouped and written into the non-volatile memory, specifically: The real-time fault data of the fault recording channel of the photovoltaic grid-connected inverter are grouped and written into the ferroelectric memory.

5. The fault recording method for photovoltaic grid-connected inverter according to any one of claims 1 to 4, characterized in that: After step S3, the method further includes: Step S4: Determine whether all the fault data in the group has been uploaded: If the judgment result is that all the fault data of the group have been uploaded, the fault recording enable flag is set to the enable flag, and then steps S2 to S4 are repeated until the recording sampling cut-off time is reached.

6. A fault diagnosis method for a photovoltaic grid-connected inverter, characterized in that: The method comprises the following steps: The fault recording method of the photovoltaic grid-connected inverter according to any one of claims 1 to 5 is used to obtain fault recording data of the photovoltaic grid-connected inverter; Extracting key features from the fault recording data to obtain extracted features; the key features include current, and / or voltage, and / or power, and / or frequency; The extracted features are diagnosed based on a machine learning model to implement fault diagnosis of the photovoltaic grid-connected inverter.

7. A DSP device, characterized in that: The device comprises: An acquisition unit, used for acquiring a fault recording enable flag; A starting unit, connected to the acquisition unit, and configured to start the photovoltaic grid-connected inverter fault recording when the fault recording enable flag bit acquired by the acquisition unit is an enable flag; A writing unit connected to the starting unit is used to write the real-time fault data of the photovoltaic grid-connected inverter fault recording channel into a non-volatile memory in groups according to the pre-configured photovoltaic grid-connected inverter fault recording channel and sampling period to obtain grouped fault data; The uploading unit is connected to the writing unit and is used to upload the grouped fault data according to the data uploading instruction to complete the fault recording of the photovoltaic grid-connected inverter.

8. The DSP device according to claim 7, characterized in that: The writing unit comprises: A setting module, used to pre-set recording channel parameters and sampling period; the recording channel is a selected channel that can reflect photovoltaic grid-connected inverter faults, and the sampling period is set according to the type of historical faults or potential faults; A division module, connected to the setting module, for dividing the non-volatile memory into N different storage areas according to preset recording channel parameters and sampling periods; A writing module, connected to the dividing module, for cyclically writing the data group into N different storage areas to obtain grouped fault data; each group of fault data includes pre-fault data, fault data and post-fault data; Wherein, N is a natural number greater than 1; A determination module, connected to the writing module, for determining the data status of the non-volatile memory according to the data written by the writing module; The first setting module is connected to the determination module and is used to set the fault recording enable flag to a disable flag when the determination module determines that all storage areas of the non-volatile memory store data and none of them are uploaded.

9. The DSP device according to claim 7, characterized in that: The uploading unit comprises: A numbering module, numbering the grouped fault data according to group numbers; An uploading module, connected to the numbering module, for uploading the grouped fault data in the order of group numbers according to a data uploading instruction; Among them, after the first group of fault data is uploaded, the non-volatile memory switches to the second group of fault data upload according to the real-time received sending completion flag; after the second group of fault data is uploaded, the non-volatile memory switches to the third group of fault data upload according to the real-time received sending completion flag; ...; after the N-1th group of fault data is uploaded, the non-volatile memory switches to the Nth group of fault data upload according to the real-time received sending completion flag, after the Nth group of fault data is uploaded, the non-volatile memory switches to the first group of fault data upload according to the real-time received sending completion flag; N is a natural number greater than 1; The second setting module is connected to the uploading module and is used to set a sending completion flag and write it into the non-volatile memory after any group of data is uploaded.

10. The DSP device according to any one of claims 7 to 9, characterized in that: The device also includes: A determination unit, used to determine whether all the fault data of the group has been uploaded; The setting unit is connected to the determination unit and is used to set the fault recording enable flag as an enable flag when the determination unit determines that all the fault data of the group has been uploaded.

11. A fault recording system for a photovoltaic grid-connected inverter, characterized in that: include: The DSP device according to any one of claims 7 to 10 is used to implement fault recording of a photovoltaic grid-connected inverter; A non-volatile memory connected to the DSP device and used to store fault recording data of the photovoltaic grid-connected inverter; Wherein, the non-volatile memory is a FRAM device; An ARM device, connected to the DSP device, for issuing a fault recording data upload instruction to the DSP device, and receiving the fault recording data sent by the DSP device; The FLASH device is connected to the ARM device and is used to store the fault recording data obtained by the ARM device from the DSP device.

Images