Inverter fault warning system and method

Abstract

The present invention discloses an inverter fault warning system and method, the system includes: an information collection device, a transmission device and a signal processing device, the information collection device is arranged at the air outlet of the inverter, and is used to collect the operation data of the inverter, and send the operation data to the transmission device; the transmission device is used to identify the received operation data, and send the identification data to the signal processing device; the signal processing device is used to perform signal processing on the received identification data, determine the characteristic signal, and compare the characteristic signal using a prediction model to determine the warning signal. By real-time collection of audio, temperature data, etc. during the operation of the inverter, it has the advantages of simple and convenient installation and low cost, and completely changes the regular inspection and cleaning mode of the traditional inverter, realizes real-time monitoring of the ventilation and heat dissipation status of the inverter, reduces the power generation loss caused by the inverter failure, and can effectively reduce the inverter failure downtime and improve the inverter availability.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an inverter fault early warning system and method. Background Art

[0002] The two core devices of a grid-connected photovoltaic power station are photovoltaic modules and inverters. Inverters are devices that convert the direct current generated by photovoltaic modules into alternating current. Usually, inverters have no redundancy. If they fail and shut down, the entire photovoltaic array will stop generating electricity, causing serious power generation losses. At present, large-scale grid-connected photovoltaic power stations are often built in deserts and uninhabited areas. Due to the large number of inverters and limited operation and maintenance personnel, untimely maintenance of inverters will seriously affect the power generation of photovoltaic power stations and even cause more serious problems. Summary of the invention

[0003] In view of this, an embodiment of the present invention provides an inverter fault early warning system and method, which solves the problem of power generation failure caused by untimely maintenance of existing inverters.

[0004] According to a first aspect, an embodiment of the present invention provides an inverter fault warning system, including: an information collection device, a transmission device, and a signal processing device, wherein:

[0005] The information collection device is arranged at the air outlet of the inverter, and is used to collect the operation data of the inverter and send the operation data to the transmission device;

[0006] The transmission device is used to identify the received operation data and send the identification data to the signal processing device;

[0007] The signal processing device is used to perform signal processing on the received identification data to determine a characteristic signal, and compare the characteristic signal using a prediction model to determine a warning signal.

[0008] The inverter fault warning system provided by the embodiment of the present invention collects audio, temperature data, etc. of the inverter in real time during operation, has the advantages of easy and convenient installation and low cost, completely changes the regular inspection and cleaning mode of traditional inverters, realizes real-time monitoring of the ventilation and heat dissipation status of the inverter, and utilizes the difference between the audio and temperature signals before and after the fault to realize the classification of inverter operation faults and intelligent identification of abnormal operation status, and early warning is used to detect potential faults in the ventilation and heat dissipation of the inverter, and switches the passive maintenance and cleaning after the fault occurs to the predictive maintenance and cleaning before the fault occurs. It can not only reduce the power generation loss caused by insufficient heat dissipation of dust accumulation and faults of the inverter, but also effectively reduce the downtime of the inverter fault, while improving the availability of the inverter, thereby improving the power generation efficiency of the photovoltaic power station.

[0009] In combination with the first aspect, in a first implementation of the first aspect, the information collection device includes: an audio collector and a temperature sensor, wherein:

[0010] The audio collector is used to collect the operating audio data of the inverter;

[0011] The temperature sensor is used to collect operating temperature data of the inverter.

[0012] In combination with the first implementation of the first aspect, in the second implementation of the first aspect, the signal processing device includes: a processor, a database, and a server, wherein:

[0013] The database is used to store the operating data and establish a prediction model based on the operating data;

[0014] The processor is used to perform signal processing on the received operation data to determine a characteristic signal;

[0015] The processor is further configured to call the corresponding prediction model according to the characteristic signal, determine a prediction result, and send the prediction result to the server;

[0016] The server determines a corresponding warning signal according to the prediction result.

[0017] In combination with the second embodiment of the first aspect, in the third embodiment of the first aspect,

[0018] The processor is used to perform sound pressure conversion on the received audio data to determine an audio characteristic signal;

[0019] The processor is used to perform discrete rate conversion on the received temperature data to determine a temperature characteristic signal.

[0020] According to a second aspect, an embodiment of the present invention provides an inverter fault early warning method, comprising:

[0021] Obtain the operating data of the inverter;

[0022] Performing signal processing on the operation data to determine a characteristic signal;

[0023] A prediction model is obtained, and the characteristic signal is compared using the prediction model to determine a warning signal.

[0024] The inverter fault warning method provided by the embodiment of the present invention collects audio data, temperature data, etc. of the inverter during operation in real time, has the advantages of easy and convenient installation and low cost, completely changes the regular inspection and cleaning mode of traditional inverters, realizes real-time monitoring of the ventilation and heat dissipation status of the inverter, and utilizes the difference between the audio and temperature signals before and after the fault to realize the classification of inverter operation faults and intelligent identification of abnormal operation status, and early warning is used to detect potential faults in the ventilation and heat dissipation of the inverter, and switches the passive maintenance and cleaning after the fault occurs to the predictive maintenance and cleaning before the fault occurs. It can not only reduce the power generation loss caused by insufficient heat dissipation of dust accumulation and faults of the inverter, but also effectively reduce the downtime of the inverter fault, while improving the availability of the inverter, thereby improving the power generation efficiency of the photovoltaic power station.

[0025] In conjunction with the second aspect, in a first implementation of the second aspect, obtaining a prediction model includes:

[0026] Obtaining a standard signal of the inverter;

[0027] Comparing the characteristic signal with the standard signal to determine a fault characteristic signal;

[0028] The prediction model is established using each of the fault characteristic signals.

[0029] In conjunction with the second aspect, in a second implementation of the second aspect, further comprising:

[0030] updating the characteristic signal using the operating data;

[0031] The prediction model is updated using the updated feature signal.

[0032] According to the third aspect, an embodiment of the present invention provides an electronic device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the inverter fault warning method described in the second aspect or any one of the embodiments of the second aspect by executing the computer instructions.

[0033] According to a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the inverter fault warning method described in the second aspect or any one of the embodiments of the second aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 is a schematic diagram of connecting main equipment of a grid-connected photovoltaic power station of a centralized inverter according to an embodiment of the present invention;

[0036] Figure 2 is an internal structural diagram of a centralized inverter according to a preferred embodiment of the present invention;

[0037] Figure 3 is a structural diagram of an inverter room and an exhaust and heat dissipation system of an inverter according to an embodiment of the present invention;

[0038] Figure 4 is a schematic diagram of an inverter fault warning system according to an embodiment of the present invention;

[0039] Figure 5 is a front view of the installation position of the acquisition device according to an embodiment of the present invention;

[0040] Figure 6 is a top view of the installation position of the acquisition device according to an embodiment of the present invention;

[0041] Figure 7 is a schematic diagram of a data transmission path according to an embodiment of the present invention;

[0042] Figure 8 is an architecture diagram of an inverter fault warning system according to an embodiment of the present invention;

[0043] Fig. 9 is a software architecture diagram of an inverter fault warning system according to an embodiment of the present invention;

[0044] Fig.10 is a flow chart of an inverter fault early warning method according to an embodiment of the present invention;

[0045] Fig.11 It is a schematic diagram of the hardware structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0047] The two core devices of a grid-connected photovoltaic power station are photovoltaic modules and inverters. The inverter is a device that converts the direct current generated by the photovoltaic module into alternating current. The inverter is divided into a centralized inverter and a string inverter. This embodiment mainly uses a centralized inverter as an example for explanation. The main equipment connection diagram of a grid-connected photovoltaic power station using a centralized inverter is as follows: Figure 1 As shown in the figure, the structure of a centralized inverter usually includes: (1) the system topology adopts a DC-AC primary power electronic device conversion full-bridge inverter and an industrial frequency isolation transformer; (2) the power device adopts a large current IGBT, the device loss is large, and forced cooling is required to ensure the equipment temperature; (3) the protection level is generally IP20, the volume is large, and it is installed vertically indoors; (4) generally single-channel maximum power point tracking (MPPT) tracking is adopted. The string voltage difference caused by abnormal shading of components, difference in string distance, difference in component factory power, inconsistent component attenuation, etc. may cause mutual interference in string MPPT voltage tracking, which will have a certain impact on the overall power generation of the photovoltaic array. Usually, the total capacity of a photovoltaic array is 1MW or more.

[0048] Usually, centralized inverters have no redundancy. If they fail, the entire photovoltaic array will stop generating electricity, causing serious power generation losses. In addition, centralized inverters are sophisticated in design and complex in structure. The internal structure of common centralized inverters is as follows: Figure 2 As shown, once a complex fault occurs, it must be inspected and repaired by professionals. The inverter-related accessories are large in size and heavy in weight. The cycle from fault discovery to fault location and then to fault elimination is long.

[0049] In addition, centralized inverters are generally installed in inverter rooms or special containers. The temperature rise of centralized inverters is higher than that of string inverters. In hot summer, it is common for the temperature in special containers or inverter rooms to exceed 50°C. The temperature inside the centralized inverter cabinet (normal operating temperature range is -30°C-55°C) is even higher and worse. High temperature reduces the life of inverter components and reduces the capacity of the entire machine. In extreme cases, there is even a risk of explosion. Every year, there are serious accidents such as explosions of centralized inverters due to heat dissipation problems.

[0050] In this embodiment, an inverter fault warning system is also provided. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0051] This embodiment provides an inverter fault warning system, wherein a heat exhaust fan is provided in the centralized inverter cabinet, and the heat generated by the inverter during operation is systematically discharged outside the inverter room through a dedicated air duct, thereby ensuring that the inverter is always within the normal temperature operating range. Figure 3 At present, the cooling fan of the centralized inverter is controlled by the dedicated temperature control module of the inverter measurement and control system. It starts automatically when the temperature reaches the set value. The heat of the inverter is discharged through a dedicated air duct. The exhaust port of the inverter cabinet is usually rectangular or circular, and a waterproof bend and insect-proof net are set.

[0052] Large-scale grid-connected photovoltaic power stations are often built in deserts and uninhabited areas. Due to the large number of inverters and limited operation and maintenance personnel, if the dust on the inverter is not cleaned in time, it is easy to cause the IGBT module and the cooling fan to accumulate too much dust, resulting in untimely heat dissipation, thereby reducing the conversion efficiency of the centralized inverter. In severe cases, the inverter will trigger a protection shutdown due to excessive operating temperature. In addition, due to the large coverage of centralized photovoltaic power stations, when the operation and maintenance personnel observe the inverter operating temperature alarm signal through the monitoring system and go to the inverter operation site for inspection and maintenance, the inverter may have stopped working, which will seriously affect the power generation of the photovoltaic power station. In addition, the existing old photovoltaic power stations generally have high-temperature tripping problems caused by the unreasonable ventilation and heat dissipation design of the centralized inverter in the early stage. The following table specifically describes the common problems of inverters and their treatment.

[0053] Table 1 Description of temperature protection function of centralized inverter

[0054]

[0055]

[0056] Table 2 Centralized inverter daily maintenance instructions

[0057]

[0058] Table 3 Common faults and solutions for centralized inverters

[0059]

[0060]

[0061] According to relevant information and actual operation observations and statistics of photovoltaic power stations, there should be no large vibration or abnormal noise when the centralized inverter is working normally. If there is any abnormality, the power should be turned off and checked in time. In addition, when the centralized inverter is operating normally, the temperature of the inverter exhaust outlet is limited to the operating temperature of the inverter. If the difference is large, the inverter should be shut down and checked in time.

[0062] The present invention discloses an inverter fault early warning system, such as Figure 4 As shown, it includes: an information collection device 1, a transmission device 3 and a signal processing device 4, wherein:

[0063] The information collection device 1 is arranged at the air outlet of the inverter 2, and is used to collect the operating data of the inverter 2 and send the operating data to the transmission device 3; the transmission device 3 is used to identify the received operating data and send the identified data to the signal processing device 4; the signal processing device 4 is used to perform signal processing on the received identified data, determine the characteristic signal, and compare the characteristic signal using the prediction model to determine the early warning signal.

[0064] Specifically, the information collection device 1 includes: an audio collector and a temperature sensor, wherein the audio collector (receiver) is used to collect the operating audio data of the inverter 2; the temperature sensor is used to collect the operating temperature data of the inverter 2. The audio collection device and the temperature sensor are installed in an embedded manner in the air outlet channel of the heat dissipation duct of the centralized inverter 2, and the installation position is as follows: Figure 5 Main view and Figure 6 As shown in the top view, the audio data and temperature data of the inverter 2 can be fixedly collected, wherein the cooling fan operation sound is collected for 30S of data each time (configurable) and 10M of data each time (configurable).

[0065] In another embodiment, the signal processing device 4 includes: a processor, a database and a server, wherein the database is used to store operation data and establish a prediction model based on the operation data; the processor is used to perform signal processing on the received operation data and determine the characteristic signal; the processor is also used to call the corresponding prediction model according to the characteristic signal, determine the prediction result, and send the prediction result to the server; the server determines the corresponding warning signal according to the prediction result. The processor is used to perform sound pressure conversion on the received audio data to determine the audio characteristic signal; the processor is used to perform discrete rate conversion on the received temperature data to determine the temperature characteristic signal.

[0066] It should be noted that the transmission paths and processing methods of the operating data of the temperature sensor and the audio collector, as well as the inverter data and ambient temperature data read by the third-party system are as follows: Figure 7 shown.

[0067] (1) The temperature sensor and audio collector collect temperature information and audio information to the intelligent processing unit, which performs sound pressure conversion and bandpass filtering on the audio to complete the collection and preliminary processing of temperature and audio data.

[0068] (2) The temperature and audio data are uploaded to the server of the photovoltaic power station (central control room), and the target audio is subjected to short-time Fourier transform, and the corresponding audio signal is processed into a frequency domain graph and an energy spectrum.

[0069] Among them, the audio frequency domain graph and energy graph of a large number of (at least 6) inverters 2 in the same station and the same batch during normal operation are collected as the standard basis; the fault information of inverter 2 and the corresponding audio frequency domain graph and energy graph before and after the fault are read through a third-party system, and the sound characteristics such as the audio frequency domain graph and energy graph before the fault are used as early warning signals. Once inverter 2 collects similar audio frequency domain graphs and energy graphs again, the fault of inverter 2 is warned in advance; such operations are repeated until the audio signal characteristics of common faults of inverter 2 are extracted, forming an inverter 2 fault warning data model. The fault information of other inverters 2 and the changes in data such as IGBT operating temperature, inverter 2 operating temperature, active power, inverter 2 conversion efficiency, etc. before and after the fault are read from the third-party system.

[0070] (3) Relevant modules are deployed on the server of the PV power station (central control room) to complete data transmission from the PV power station to the cloud server.

[0071] (4) Call cloud services to store the processed data on the data platform, and perform deep learning of the corresponding algorithms and model updates based on the platform.

[0072] In the embodiment of the present invention, the system principle and system architecture adopted are as follows Figure 8 As shown: the acquisition end (terminal): consists of an intelligent processing unit, a temperature sensor and an audio collector. The intelligent processing unit collects real-time data from the temperature sensor and the audio collector, collects front-end data according to the plan and performs specific data processing (such as temperature discrete rate, bandpass filtering, etc.), and transmits the inverter 2 cooling duct temperature data and inverter 2 operating audio data to the photovoltaic power station (central control room) server through the photovoltaic power station communication ring network.

[0073] Server: includes algorithm management system, database service system, and WEB server system. The algorithm management system manages the scheduling of audio processing algorithms and post-algorithm processing. After receiving audio data, the algorithm management system calls the corresponding algorithm for processing. The audio data predicts common faults in advance based on the corresponding fault data model. Subsequent audio data calls the corresponding algorithm to analyze the waveform, spectrum, time domain diagram, match the model, and output the audio frequency domain diagram and energy diagram before and after the fault for comparison.

[0074] As massive amounts of fault data enter the database, fault big data is formed, and these data are used again for further modeling and deep learning to form a more compact model database.

[0075] The audio collector is an audio collection sensor; the intelligent processing unit is responsible for audio data collection, preprocessing and sending. The centralized photovoltaic inverter 2 dust accumulation fault warning system is mainly based on the collected inverter 2 operation sound data, temperature data, inverter 2 process operation related data (inverter 2 operation temperature, power and ambient temperature, etc.), other inverter 2 similar data, combined with time domain and frequency domain analysis methods for feature engineering, combined with machine learning algorithms to achieve the detection and classification of inverter 2 operation abnormalities.

[0076] like Fig. 9 As shown in the software architecture, the centralized photovoltaic inverter 2 dust accumulation fault warning system is responsible for the interaction with the inverter 2 operating information, audio data storage and processing, audio data real-time display, inverter 2 dust accumulation degree prediction (when the inverter 2 fan is seriously dusty, although the inverter 2 cannot report the fan operation failure, the sound of the inverter 2 operation at this time is lower than the recorded historical normal operation sound of the inverter 2, and the audio frequency domain diagram and energy spectrum change significantly. In addition, the inverter 2 is seriously dusty and the fan heat dissipation capacity decreases. The temperature of the inverter 2 increases, causing its conversion efficiency to decrease) and equipment fault warning push and other tasks, and provides a user-friendly operation interface. It is a terminal for machine learning and algorithm configuration.

[0077] Through the top-level design of the software architecture, the software can control the audio collector to collect the corresponding audio data according to different task requirements and continuously improve the system functions through the continuous upgrading and optimization of the fault identification module. At the same time, a corresponding database is established to store data for self-optimization of the system. The blade audio and temperature monitoring system is used to monitor the health status of inverter 2. In the audio display interface, you can not only click on the fixed audio playback of each inverter 2 in the photovoltaic power station, but also listen to the sound of inverter 2 running. You can also view the audio processing results: frequency domain diagram and energy diagram.

[0078] The inverter fault warning system provided by the embodiment of the present invention collects audio, temperature data, etc. of the inverter in real time during operation, has the advantages of easy and convenient installation and low cost, completely changes the regular inspection and cleaning mode of traditional inverters, realizes real-time monitoring of the ventilation and heat dissipation status of the inverter, and utilizes the difference between the audio and temperature signals before and after the fault to realize the classification of inverter operation faults and intelligent identification of abnormal operation status, and early warning is used to detect potential faults in the ventilation and heat dissipation of the inverter, and switches the passive maintenance and cleaning after the fault occurs to the predictive maintenance and cleaning before the fault occurs. It can not only reduce the power generation loss caused by insufficient heat dissipation of dust accumulation and faults of the inverter, but also effectively reduce the downtime of the inverter fault, while improving the availability of the inverter, thereby improving the power generation efficiency of the photovoltaic power station.

[0079] In this embodiment, an inverter fault early warning method is provided, which can be used in electronic devices such as computers, mobile phones, tablet computers, etc. Fig.10 is a flow chart of an inverter fault early warning method according to an embodiment of the present invention. Fig.10 As shown, the process includes the following steps:

[0080] S11, obtaining the operating data of the inverter.

[0081] In this embodiment, the operating audio data of the inverter is collected by an audio collector, and the operating temperature data of the inverter is collected by a temperature sensor, so as to ensure the real-time and accuracy of the data.

[0082] S12, performing signal processing on the operation data to determine a characteristic signal.

[0083] In this embodiment, the main factors to be considered are sampling frequency, data integrity, inverter operating conditions, etc. Based on the sound data of the inverter operation, feature extraction is performed through time domain and frequency domain analysis. Feature engineering is performed by combining time domain and frequency domain analysis methods.

[0084] S13, obtaining a prediction model, and using the prediction model to compare characteristic signals to determine a warning signal.

[0085] Wherein, obtaining the prediction model includes: obtaining a standard signal of the inverter; comparing the characteristic signal with the standard signal to determine a fault characteristic signal; and establishing a prediction model using each fault characteristic signal.

[0086] The prediction models include: fault-free model and faulty model. Fault-free model: the dust accumulation degree of the inverter is estimated based on the difference in the operating sound of the inverters in the same power station and the same batch; faulty model: the machine learning classification model is trained based on the time domain and frequency domain features, such as the audio feature extraction and judgment before and after the common faults of the inverter occur.

[0087] The audio analysis of the centralized photovoltaic inverter dust accumulation fault warning system converts the audio data in the time domain into a spectrum in the frequency domain, identifies spectrum anomalies through algorithms, determines the impending inverter fault and generates a warning. Currently, several types of inverter faults can be extracted and identified. Table 4 shows the inverter operation sound feature extraction data.

[0088] Table 4 Inverter operation sound feature extraction table

[0089]

[0090] The inverter fault warning method provided in this embodiment collects audio and temperature data of the inverter during operation in real time, is easy to install and has low cost, completely changes the regular inspection and cleaning mode of traditional inverters, realizes real-time monitoring of the ventilation and heat dissipation status of the inverter, and uses the differences in audio and temperature signals before and after the fault to realize the classification of inverter operation faults and intelligent identification of abnormal operation status, and early warning is used to detect potential faults in the ventilation and heat dissipation of the inverter, and switches the passive maintenance and cleaning after the fault occurs to the pre-judgment maintenance and cleaning before the fault occurs. It can not only reduce the power generation loss caused by insufficient heat dissipation of dust accumulation and faults of the inverter, but also effectively reduce the downtime of the inverter fault, while improving the availability of the inverter, thereby improving the power generation efficiency of the photovoltaic power station.

[0091] The present invention also provides an electronic device. Fig.11 , Fig.11 is a schematic diagram of the structure of an electronic device provided by an optional embodiment of the present invention, such as Fig.11 As shown, the electronic device may include: at least one processor 601, such as a CPU (Central Processing Unit), at least one communication interface 603, a memory 604, and at least one communication bus 602. The communication bus 602 is used to realize the connection and communication between these components. The communication interface 603 may include a display screen (Display) and a keyboard (Keyboard), and the optional communication interface 603 may also include a standard wired interface and a wireless interface. The memory 604 may be a high-speed RAM memory (Random Access Memory) or a non-volatile memory (non-volatile memory), such as at least one disk storage. The memory 604 may optionally be at least one storage device located away from the aforementioned processor 601. The processor 601 may be combined with Figure 5 In the described device, the memory 604 stores an application program, and the processor 601 calls the program code stored in the memory 604 to execute any of the above method steps.

[0092] The communication bus 602 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The communication bus 602 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.11 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0093] Among them, the memory 604 may include a volatile memory (English: volatile memory), such as a random access memory (English: random-access memory, abbreviated: RAM); the memory may also include a non-volatile memory (English: non-volatile memory), such as a flash memory (English: flash memory), a hard disk drive (English: hard disk drive, abbreviated: HDD) or a solid-state drive (English: solid-state drive, abbreviated: SSD); the memory 604 may also include a combination of the above types of memory.

[0094] The processor 601 may be a central processing unit (CPU), a network processor (NP), or a combination of a CPU and a NP.

[0095] The processor 601 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0096] Optionally, the memory 604 is also used to store program instructions. The processor 601 can call the program instructions to implement the inverter fault early warning method as shown in the embodiment of the present application.

[0097] The embodiment of the present invention further provides a non-transitory computer storage medium, which stores computer executable instructions, and the computer executable instructions can execute the inverter fault warning method in any of the above method embodiments. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory (Flash Memory), a hard disk (HDD) or a solid-state drive (SSD), etc.; the storage medium can also include a combination of the above types of memory.

[0098] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. An inverter fault early warning system, characterized in that: include: Information collection device, transmission device and signal processing device, wherein: The information collection device is arranged at the air outlet of the inverter, and is used to collect the operation data of the inverter and send the operation data to the transmission device; The transmission device is used to identify the received operation data and send the identification data to the signal processing device; The signal processing device is used to perform signal processing on the received identification data to determine a characteristic signal, and compare the characteristic signal using a prediction model to determine a warning signal; The information collection device includes: an audio collector and a temperature sensor, wherein: The audio collector is used to collect the operating audio data of the inverter; The temperature sensor is used to collect operating temperature data of the inverter; The signal processing device includes: a processor, a database and a server, wherein: The database is used to store the operating data and establish a prediction model based on the operating data; The processor is used to perform signal processing on the received operation data to determine the characteristic signal, including: performing sound pressure conversion on the received audio data to determine the audio characteristic signal; performing discrete rate conversion on the received temperature data to determine the temperature characteristic signal; The processor is further configured to call the corresponding prediction model according to the characteristic signal, determine a prediction result, and send the prediction result to the server; The server determines a corresponding warning signal according to the prediction result; The prediction model includes a fault-free model and a faulty model; the fault-free model estimates the dust accumulation degree of the inverter based on the difference in the operating sound of the inverters in the same power station and the same batch; the faulty model trains the machine learning classification model based on time domain and frequency domain features, including audio feature extraction and judgment before and after common inverter faults occur; Centralized photovoltaic inverter dust accumulation fault warning includes: converting audio data in the time domain into a spectrum in the frequency domain, identifying spectrum anomalies through algorithms, determining the impending inverter failure and generating a warning; The audio characteristic signal determination method comprises: The audio frequency domain graphs and energy graphs of a large number of inverters in the same station and the same batch during normal operation are collected as the standard basis; the fault information of the inverter and the corresponding audio frequency domain graphs and energy graphs before and after the fault are read through a third-party system, and the sound characteristics of the audio frequency domain graph and energy graph before the fault are used as early warning signals. Once the inverter collects similar audio frequency domain graphs and energy graphs again, the inverter fault is warned in advance; this type of operation is repeated until the audio signal characteristics of common inverter faults are extracted, forming an inverter fault warning data model.

2. An inverter fault early warning method, characterized in that: include: Obtain the operating data of the inverter; The operation data is subjected to signal processing to determine a characteristic signal, including: performing sound pressure conversion on the received audio data to determine an audio characteristic signal; performing discrete rate conversion on the received temperature data to determine a temperature characteristic signal; the method for determining the audio characteristic signal includes: collecting audio frequency domain graphs and energy graphs of a large number of inverters of the same station and the same batch during normal operation as a standard basis; reading the fault information of the inverter and the corresponding audio frequency domain graphs and energy graphs before and after the fault through a third-party system, and using the sound characteristics of the audio frequency domain graph and energy graph before the fault occurs as an early warning signal. Once the inverter collects similar audio frequency domain graphs and energy graphs again, an early warning of the inverter fault is given; such operations are repeated until the audio signal characteristics of common faults of the inverter are extracted, thereby forming an inverter fault early warning data model; A prediction model is obtained, and the characteristic signal is compared with the prediction model to determine a warning signal; the prediction model includes a fault-free model and a faulty model; the fault-free model estimates the degree of dust accumulation of the inverter based on the difference in operating sounds of the same power station and the same batch of inverters; the faulty model trains the machine learning classification model based on time domain and frequency domain features, including audio feature extraction and judgment before and after common inverter faults occur; the centralized photovoltaic inverter dust accumulation fault warning includes: converting the time domain audio data into a frequency spectrum diagram in the frequency domain, identifying spectrum anomalies through algorithms, judging the impending fault of the inverter and generating a warning.

3. An electronic device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the inverter fault warning method according to claim 2 by executing the computer instructions.

4. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the inverter fault early warning method according to claim 2.

Images