Photovoltaic panel installation robot monitoring platform fault judgment method and device and medium

Abstract

The invention discloses a photovoltaic panel installation robot monitoring platform fault judgment method and device and a medium, and relates to the technical field of intelligent operation and maintenance of photovoltaic power generation equipment.The method comprises the steps that firstly, a data quality evaluation matrix is established for received multivariate data, low-quality data is removed according to a threshold value, and intelligent screening and optimization of the multivariate data are achieved; then carrying out joint feature extraction on the data passing the quality evaluation, and constructing a multi-dimensional feature vector; then constructing a Lora fine tuning model; and finally, outputting system state information and a fault alarm signal according to the fine-tuned model. According to the invention, multivariate data fusion can be realized, the running state of the robot can be monitored in real time, various fault features can be accurately identified, and a reliable basis is provided for operation and maintenance decisions. And through a highly integrated technical means, comprehensive monitoring of the health condition of the photovoltaic panel installation robot is realized, so that safe and stable operation of equipment is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent operation and maintenance of photovoltaic power generation equipment, and in particular to a fault judgment method, device and medium for a photovoltaic panel installation robot monitoring platform. Background Art

[0002] The statements in this section merely provide background information related to the present disclosure and may not constitute prior art.

[0003] With the promotion and application of photovoltaic power generation technology, the intelligent operation and maintenance of photovoltaic panel installation robots face severe challenges. Existing fault diagnosis technologies have the following significant defects: traditional analysis methods are difficult to effectively process and analyze the complex data streams generated by the collaborative work of multiple systems, resulting in insufficient extraction of key fault features and a high misjudgment rate. In particular, for the collaborative monitoring of key components such as the main controller, range extender, motor drive and battery system, the existing system has technical bottlenecks in the data fusion processing level. In addition, conventional monitoring solutions are limited by the efficiency of data processing algorithms. In terms of real-time response speed and diagnostic accuracy, it is difficult to meet the precise operation and maintenance requirements in complex field operating environments, and cannot effectively support preventive maintenance decisions for equipment. Key technical problems that need to be solved in the current technical system include intelligent processing of multi-source heterogeneous data, joint extraction of multi-dimensional features, and construction of real-time fault diagnosis models. Summary of the Invention

[0004] The present invention aims to address the problems existing in the prior art by providing a fault diagnosis method, device, and medium for a photovoltaic panel installation robot monitoring platform. These methods, which enable multi-dimensional data fusion, enable real-time monitoring of the robot's operating status, accurately identify various fault characteristics, and provide a reliable basis for operational and maintenance decision-making. Through highly integrated technical means, comprehensive monitoring of the photovoltaic panel installation robot's health status is achieved, ensuring the safe and stable operation of the equipment.

[0005] The technical solutions of the present invention are as follows:

[0006] A photovoltaic panel installation robot monitoring platform fault diagnosis method, comprising:

[0007] Step S1: receiving multi-dimensional data periodically sent by the photovoltaic panel installation robot network system;

[0008] Step S2: Establish a data quality assessment matrix for the received multivariate data, and eliminate low-quality data based on the threshold value to achieve intelligent screening and optimization of multivariate data;

[0009] Step S3: Perform joint feature extraction on the data that pass the quality assessment and construct a multi-dimensional feature vector;

[0010] Step S4: construct the Lora fine-tuning model;

[0011] Step S5: Output system status information and fault alarm signals according to the fine-tuned model.

[0012] Furthermore, the multivariate data includes: main controller data, range extender ECU data, motor driver MCU data and battery system data.

[0013] Furthermore, the quality assessment matrix is ​​implemented through weighted calculation of completeness, timeliness and accuracy.

[0014] Furthermore, the quality assessment matrix is ​​calculated as follows:

[0015] Q = α·Completeness + β·Timeliness + γ·Accuracy

[0016] Q represents the comprehensive score of data quality;

[0017] α, β, and γ represent weights, α+β+γ=1;

[0018] Completeness represents the completeness score of the data;

[0019] Timeliness indicates the delay time score of the data;

[0020] Accuracy represents the rationality score of the data, which can be obtained based on the check digits of the data.

[0021] Furthermore, the features extracted in step S3 include: time domain features and frequency domain features.

[0022] Furthermore, the time domain features include: mean μ, variance σ of the data 2 and peak-to-peak value x pp .

[0023] Furthermore, the frequency domain characteristics include: the dominant frequency amplitude A peak Sum band energy ratio E band .

[0024] Furthermore, the step S4 includes:

[0025] Perform low-rank decomposition on the pre-trained model weight matrix.

[0026] The present invention also proposes a photovoltaic panel installation robot monitoring platform fault judgment device, including: a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the computer program, the steps of the photovoltaic panel installation robot monitoring platform fault judgment method as described above are implemented.

[0027] The present invention also proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of the above-mentioned photovoltaic panel installation robot monitoring platform fault judgment method.

[0028] Compared with the existing technology, the beneficial effects of the present invention are:

[0029] 1. The present invention significantly improves the fault diagnosis efficiency of photovoltaic panel installation robots through multivariate data fusion and intelligent analysis. By integrating the operating data of the main controller, range extender, motor driver and battery system, the system breaks through the limitations of traditional single-dimensional monitoring. Combined with an innovative data quality assessment matrix, it dynamically optimizes the data screening process to ensure that the confidence of the input data is improved. In the feature extraction stage, a collaborative strategy of time domain statistical analysis and frequency domain transformation is adopted. Transient anomalies are captured through time domain features such as mean, variance, and peak-to-peak value. At the same time, Fourier transform is used to extract the frequency band energy distribution, which improves the recognition accuracy of periodic faults such as bearing wear and achieves comprehensive coverage of multi-dimensional fault features.

[0030] 2. The present invention innovatively introduces the Lora low-rank decomposition technology to perform lightweight fine-tuning on the pre-trained model, compresses the number of parameters through weight matrix decomposition, and shortens the fault diagnosis response time while retaining the deep model feature learning ability. This technology can not only accurately identify multiple types of faults such as battery voltage fluctuations and abnormal motor vibrations, with a lower misjudgment rate than traditional methods, but also reduce unplanned downtime through real-time status monitoring. In addition, through timeliness evaluation and frequency band energy dynamic analysis, the system effectively overcomes signal delays and electromagnetic interference in field operations, improves diagnostic stability in complex environments, saves comprehensive operation and maintenance costs, and provides an efficient and reliable solution for the intelligent operation and maintenance of photovoltaic equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a flow chart of a fault diagnosis method for a photovoltaic panel installation robot monitoring platform. DETAILED DESCRIPTION

[0032] It should be noted that relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

[0033] The features and performance of the present invention are further described in detail below with reference to the embodiments.

[0034] Example 1

[0035] See also Figure 1 A photovoltaic panel installation robot monitoring platform fault judgment method specifically includes the following steps:

[0036] Step S1: receiving multi-dimensional data periodically sent by the photovoltaic panel installation robot network system;

[0037] Step S2: Establish a data quality assessment matrix for the received multivariate data, and eliminate low-quality data based on the threshold value to achieve intelligent screening and optimization of multivariate data;

[0038] Step S3: Joint feature extraction is performed on the data that has passed the quality assessment to construct a multi-dimensional feature vector. That is, feature engineering is performed on the data. Through time domain analysis and frequency domain transformation, key features are extracted from the original data to construct a multi-dimensional feature vector, providing high-information input data for the subsequent fault diagnosis model.

[0039] Step S4: construct the Lora fine-tuning model;

[0040] Step S5: Output system status information and fault alarm signals according to the fine-tuned model.

[0041] In this embodiment, specifically, the multivariate data includes: main controller data, range extender ECU data, motor driver MCU data and battery system data.

[0042] In this embodiment, specifically, the quality assessment matrix is ​​implemented through weighted calculation of completeness, timeliness and accuracy.

[0043] In this embodiment, specifically, the quality assessment matrix is ​​calculated as follows:

[0044] Q = α·Completeness + β·Timeliness + γ·Accuracy

[0045] Q represents the comprehensive score of data quality;

[0046] α, β, and γ represent weights, which can be adjusted according to actual conditions, α+β+γ=1;

[0047] Completeness represents the completeness score of the data;

[0048] Timeliness indicates the delay time score of the data;

[0049] Accuracy indicates the plausibility score of the data;

[0050] It should be noted that completeness, timeliness and accuracy can be calculated using common data processing methods; for example:

[0051] Completeness: Detects the degree of data missing, evaluates field integrity and sampling continuity; for example, based on the number of data bytes, if the original data is 100 bytes and 100 bytes are collected, the completeness score is 100; if 30 bytes are collected, the completeness score is 30;

[0052] Timeliness: Directly calculating the data delay time is timeliness;

[0053] Accuracy: Data rationality is verified through threshold verification and logical rules. For example, based on the data's check digit, data can generally be either accurate or incorrect. Therefore, if the check digit passes, the value is 100, and if it fails, the value is 0.

[0054] Of course, other strategies may also be used to calculate the completeness, timeliness, and accuracy scores, which are not limited in the present invention.

[0055] In this embodiment, specifically, the features extracted in step S3 include: time domain features and frequency domain features.

[0056] In this embodiment, specifically, the time domain features include: mean μ, variance σ of the data 2 and peak-to-peak value x pp .

[0057] In this embodiment, specifically, the frequency domain features include: the dominant frequency amplitude A peak Sum band energy ratio E band .

[0058] In this embodiment, it should be noted that the mean μ is calculated using the following formula:

[0059]

[0060] in:

[0061] N represents the total number of data;

[0062] x i Represents the value of the data itself.

[0063] In this embodiment, it should be noted that the variance σ 2 Calculated by the following formula:

[0064]

[0065] In this embodiment, it should be noted that the peak-to-peak value x pp Calculated by the following formula:

[0066] x pp =max(x i )-min(x i )

[0067] In this embodiment, it should be noted that the present invention uses Fourier transform to extract frequency domain features, which is mainly used for identifying characteristic frequencies of motor bearing faults. The specific implementation method is as follows:

[0068] 1. Use Fourier transform to extract frequency domain features. Fourier transform uses the formula:

[0069]

[0070] 2. Calculate the dominant frequency amplitude A peak :

[0071] A peak =max(|X(f)|)

[0072] 3. Calculate the frequency band energy ratio E band :

[0073]

[0074] 4. Construct the feature matrix and generate the final feature vector:

[0075] F=[μ,σ 2 ,x pp ,A peak ,E band ] T

[0076] In this embodiment, it should be noted that some other eigenvalues ​​may be introduced into the multi-dimensional eigenvector, which are selected according to actual conditions.

[0077] In this embodiment, specifically, step S4 includes:

[0078] For the pre-trained model weight matrix W∈R m x n Perform low-rank decomposition;

[0079] W=W0+ΔW=W0+A·B T

[0080] in:

[0081] W0 is the initial weight matrix of the pre-trained model;

[0082] ΔW is the low-rank update matrix;

[0083] A∈R m×r and A∈R r×n is a low-rank decomposition matrix, r is a low-rank dimension (usually r< <m,n)。

[0084] Example 2

[0085] Example 2 also proposes a photovoltaic panel installation robot monitoring platform fault judgment device, including: a memory, a processor, and a computer program stored in the memory and runnable on the processor, wherein the processor implements the steps of the photovoltaic panel installation robot monitoring platform fault judgment method as described above when executing the computer program; preferably, the computer program can be run on a terminal device, such as a personal computer.

[0086] Example 3

[0087] Example 3 also proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of the fault judgment method of a photovoltaic panel installation robot monitoring platform as described above; however, the device of the present invention is not limited to this. In this document, the readable storage medium can be any tangible medium that contains or stores a program, which can be used by or in combination with an instruction execution system, device or component.

[0088] The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0089] The computer-readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, wherein the readable program code is carried. The data signal propagated may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or component. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination thereof.

[0090] The program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, and the like, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a stand-alone software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0091] The above-described embodiments merely represent specific implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of protection of the present application. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the technical concept of the present application, and all such variations and improvements fall within the scope of protection of the present application.

[0092] This background section is provided to generally present the context of the invention, and the work of the presently named inventors, the work to the extent described in this background section, and aspects of the description in this section that did not constitute prior art at the time of filing are neither explicitly nor implicitly admitted to be prior art to the present invention.

Claims

1. A photovoltaic panel installation robot monitoring platform fault judgment method, characterized in that: include: Step S1: receiving multi-dimensional data periodically sent by the photovoltaic panel installation robot network system; Step S2: Establish a data quality assessment matrix for the received multivariate data, and eliminate low-quality data based on the threshold value to achieve intelligent screening and optimization of multivariate data; Step S3: Perform joint feature extraction on the data that pass the quality assessment and construct a multi-dimensional feature vector; Step S4: construct the Lora fine-tuning model; Step S5: Output system status information and fault alarm signals according to the fine-tuned model.

2. A photovoltaic panel installation robot monitoring platform fault judgment method according to claim 1, characterized in that: The multivariate data includes: main controller data, range extender ECU data, motor driver MCU data and battery system data.

3. A photovoltaic panel installation robot monitoring platform fault judgment method according to claim 1, characterized in that: The quality assessment matrix is ​​implemented through weighted calculation of completeness, timeliness and accuracy.

4. A photovoltaic panel installation robot monitoring platform fault judgment method according to claim 3, characterized in that: The quality assessment matrix is ​​calculated as follows: Q = α·Completeness + β·Timeliness + γ·Accuracy Q represents the comprehensive score of data quality; α, β, and γ represent weights, α+β+γ=1; Completeness represents the completeness score of the data; Timeliness indicates the delay time score of the data; Accuracy indicates the plausibility score of the data.

5. A photovoltaic panel installation robot monitoring platform fault judgment method according to claim 1, characterized in that: The features extracted in step S3 include: time domain features and frequency domain features.

6. A photovoltaic panel installation robot monitoring platform fault judgment method according to claim 5, characterized in that: The time domain features include: mean μ, variance σ of the data 2 and peak-to-peak value x pp .

7. A photovoltaic panel installation robot monitoring platform fault judgment method according to claim 6, characterized in that: The frequency domain characteristics include: the dominant frequency amplitude A peak Sum band energy ratio E band .

8. The photovoltaic panel installation robot monitoring platform fault judgment method according to claim 1, characterized in that: The step S4 comprises: Perform low-rank decomposition on the pre-trained model weight matrix.

9. A photovoltaic panel installation robot monitoring platform fault diagnosis device, comprising: A memory, a processor, and a computer program stored in the memory and runnable on the processor, characterized in that when the processor executes the computer program, the steps of a fault judgment method for a photovoltaic panel installation robot monitoring platform as described in any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of a fault judgment method for a photovoltaic panel installation robot monitoring platform as described in any one of claims 1 to 8 are implemented.

Images