CN114443605B

CN114443605B - Information analysis method and system for overwater photovoltaic system

Info

Publication number: CN114443605B
Application number: CN202210340667.0A
Authority: CN
Inventors: 于俊; 王建明; 介雷; 周静; 刘勇
Original assignee: Das Solar Co Ltd
Current assignee: Yidao New Energy Technology Co ltd
Priority date: 2022-04-02
Filing date: 2022-04-02
Publication date: 2022-07-29
Anticipated expiration: 2042-04-02
Also published as: CN114443605A

Abstract

The information analysis method and the system for the overwater photovoltaic system of the invention perform downsampling operation on the received first overwater photovoltaic operation log queue to obtain a first downsampling description information queue pointed by the first overwater photovoltaic operation log queue, perform derivative analysis by combining the first downsampling description information queue to obtain a second downsampling description information queue for completing derivative analysis, perform log reduction operation on the second downsampling description information queue to obtain a first derivative operation log queue pointed by the second downsampling description information queue, perform derivative analysis on the first downsampling description information queue obtained by downsampling, obviously reduce the information scale faced by the derivative analysis process, effectively reduce the resource overhead of derivative analysis, and improve the prospective analysis efficiency of the operation state of the overwater photovoltaic system, so as to quickly and accurately determine a first derived running log queue reflecting the possible subsequent running state of the overwater photovoltaic system.

Description

Information analysis method and system for overwater photovoltaic system

Technical Field

The invention relates to the technical field of overwater photovoltaic, in particular to an information analysis method and system for an overwater photovoltaic system.

Background

Photovoltaic system on water refers to the photovoltaic power generation system of construction on the surface of water. Different from a ground power generation system, the water photovoltaic system does not occupy land resources, and is very suitable for areas with rich water resources and deficient land resources. In addition, the water body has a cooling effect on the photovoltaic equipment, so that the temperature rise of the surface of the equipment can be inhibited, the radiation from the surface of the water body can be reduced, and the overall power generation amount is slightly higher than that of a ground power generation system under the same condition. Along with photovoltaic system's on water continuous application, how to realize photovoltaic system's on water running state analysis on a high efficiency basis is the key point of attacking of present photovoltaic technology on water.

Disclosure of Invention

According to an aspect of the present invention, there is provided an information analysis method for an above-water photovoltaic system, including: performing downsampling operation on the received first photovoltaic operation log queue on the water to obtain a first downsampling description information queue pointed by the first photovoltaic operation log queue on the water; performing a derivative analysis in conjunction with the first downsampling descriptor queue to obtain a second downsampling descriptor queue for which the derivative analysis is completed; and performing log reduction operation on the second downsampling description information queue to obtain a first derivative running log queue pointed by the second downsampling description information queue.

According to the design, the down-sampling operation is performed on the received first photovoltaic operation log queue on the water to obtain a first down-sampling description information queue pointed by the first photovoltaic operation log queue on the water, the derivative analysis is performed by combining the first down-sampling description information queue to obtain a second down-sampling description information queue subjected to derivative analysis, and the log reduction operation is performed on the second down-sampling description information queue to obtain a first derivative operation log queue pointed by the second down-sampling description information queue, so that the derivative analysis is performed on the first down-sampling description information queue obtained through down-sampling, the scale of information faced by the derivative analysis process is obviously reduced, the resource overhead of derivative analysis can be effectively reduced, the prospective analysis efficiency of the operation state of the photovoltaic system on the water is improved, and the first derivative operation state reflecting the possible subsequent occurrence of the photovoltaic system on the water can be determined as quickly and accurately as possible And generating a running log queue.

For some possible embodiments, after the obtaining the first derived run log queue pointed to by the second downsampled description information queue, the method further comprises: performing derivative analysis in combination with at least part of the water photovoltaic operation logs in the first water photovoltaic operation log queue to obtain a second derivative operation log queue for completing derivative analysis; and combining the first derived running log queue and the second derived running log queue to obtain a third derived running log queue for completing derived analysis.

By means of the design, the photovoltaic operation logs on at least part of the water in the first overwater photovoltaic operation log queue are combined to perform derivative analysis to obtain a second derivative operation log queue with derivative analysis completed, and the first derivative operation log queue and the second derivative operation log queue are combined to obtain a third derivative operation log queue with derivative analysis completed, so that operation log derivative analysis can be performed based on derivative operation log information as complete as possible, and the precision of derivative analysis results can be further improved.

For some possible embodiments, the performing a downsampling operation on the received first photovoltaic operation log queue on water to obtain a first downsampling description information queue pointed to by the first photovoltaic operation log queue on water includes: performing set significance operation label mapping on a received first underwater photovoltaic operation log queue to obtain a first simplified description information queue pointed by the first underwater photovoltaic operation log queue; performing unconstrained updating on the first simplified description information queue to obtain a first downsampling description information queue pointed to by the first above-water photovoltaic operation log queue.

By means of the design, the set significance operation label mapping and the non-constraint updating are carried out on the first overwater photovoltaic operation log queue, the characteristic size is reduced, and a first lower sampling description information queue with a smaller characteristic size is obtained and used for subsequent derivative analysis processing. By performing the set significance operation label mapping and the non-constraint updating on the first overwater photovoltaic operation log queue to perform the down-sampling operation, compared with the related art, the operation difficulty can be reduced. By performing the set significance operation label mapping and the unconstrained updating on the first overwater photovoltaic operation log queue to perform the downsampling operation, the processing timeliness can be improved.

For some possible embodiments, the performing derivative analysis in conjunction with the first downsampling descriptor queue to obtain a second downsampling descriptor queue for which derivative analysis is completed includes: and loading the first downsampling description information queue to a first staged AI model, and obtaining a second downsampling description information queue pointed by the first downsampling description information queue by utilizing the first staged AI model for derivative analysis.

By means of the design, the first staged AI model is combined with the first downsampling description information queue to perform derivative analysis so as to obtain a second downsampling description information queue for completing derivative analysis, and therefore derivative analysis precision and derivative analysis timeliness can be improved.

For some possible embodiments, the performing a log reduction operation on the second downsampled description information queue to obtain a first derived running log queue pointed to by the second downsampled description information queue includes: performing log reduction operation on the second downsampling description information queue to obtain a second simplified description information queue pointed by the second downsampling description information queue; and performing significance operation label restoration on the second simplified description information queue to obtain a first derivative operation log queue pointed by the second downsampling description information queue.

By means of the design, the log reduction operation is performed on the second downsampling description information queue to obtain a second simplified description information queue, the set significance operation label reduction is performed on the second simplified description information queue, and the downsampling description information in the second downsampling description information queue can be respectively translated into the derivative operation logs which are consistent with the derivative operation log information distribution structure in the first overwater photovoltaic operation log queue.

For some possible embodiments, the performing a log reduction operation on the second downsampled descriptor queue to obtain a second simplified descriptor queue pointed to by the second downsampled descriptor queue includes: and executing cyclic processing on the second downsampling description information queue by adopting a trigger algorithm of a dynamic judgment value to obtain a second simplified description information queue pointed by the second downsampling description information queue.

By adopting the design, the second simplified description information queue pointed by the second downsampling description information queue can be obtained in time by executing cyclic processing on the second downsampling description information queue by adopting a triggering algorithm of a dynamic judgment value, so that the timeliness of running log derivative analysis can be improved.

For some possible embodiments, the performing a derivative analysis in conjunction with at least some of the photovoltaic logbooks on the water in the first queue of photovoltaic logbooks to obtain a second queue of derivative logbooks that completes the derivative analysis includes: and loading at least part of the water photovoltaic operation logs in the first water photovoltaic operation log queue to a second staged AI model, and performing derivative analysis by using the second staged AI model to obtain a second derivative operation log queue pointed by the at least part of the water photovoltaic operation logs.

By means of the design, the photovoltaic operation logs on at least part of the water in the first overwater photovoltaic operation log queue are processed through the second stage AI model to obtain a second derivative operation log queue, and therefore high-quality derivative operation log information required by subsequent operation log derivative analysis can be obtained. Therefore, the final derivative analysis result is obtained based on the second derivative operation log queue and the first derivative operation log queue, and the accuracy of the log derivative analysis and the operation state prediction can be further improved.

For some possible embodiments, the at least part of the above-water photovoltaic operation logs includes X above-water photovoltaic operation logs currently received in the first above-water photovoltaic operation log queue, where X is a positive integer, and the number of the above-water photovoltaic operation logs in the first above-water photovoltaic operation log queue is not less than X.

By the design, the currently received X overwater photovoltaic operation logs in the first overwater photovoltaic operation log queue are adopted to execute derivative analysis, and the accuracy of log derivative analysis and operation state prediction is improved.

For some possible embodiments, the combining the first derived run log queue and the second derived run log queue to obtain a third derived run log queue that completes the derived analysis includes: performing significance running label mining on the first derivative running log queue to obtain a first significance running label pointed by the first derivative running log queue; performing significance running label mining on the second derivative running log queue to obtain a second significance running label pointed by the second derivative running log queue; performing a first stitching process in conjunction with the first salient running label and the second salient running label to obtain a first stitched running label; and combining the first spliced running label to obtain a third derivative running log queue for finishing derivative analysis.

By means of the design, the first significance running label extracted from the first derivative running log queue and the second significance running label extracted from the second derivative running log queue are combined, running log derivative analysis can be performed based on derivative running log information which is as complete as possible, and therefore the accuracy of log derivative analysis and running state prediction can be improved.

For some possible embodiments, the combining the first spliced running label to obtain a third derived running log queue for which the derived analysis is completed includes: performing a linear regression operation on the first spliced running label to obtain a target significance running label; and combining the first spliced running label and the target significance running label to obtain a third derivative running log queue for completing derivative analysis.

By means of the design, the target significance running label is obtained by performing linear regression operation on the first spliced running label, and the third derivative running log queue for completing derivative analysis is obtained by combining the first spliced running label and the target significance running label, so that disturbance of an AI model can be avoided as much as possible, and the precision of log derivative analysis and running state prediction is further improved. Further, aiming at the debugging process of the AI model, by executing the linear regression operation, the convergence of the AI model can be realized as fast as possible so as to ensure that the AI model quickly reaches a stable state.

For some possible embodiments, the first spliced operational label includes a plurality of photovoltaic power generation environmental themes; the combining the first spliced running label and the target saliency running label to obtain a third derived running log queue completing the derived analysis comprises: performing second splicing processing on the first spliced running label of the last photovoltaic power generation environment theme to obtain a second spliced running label; and combining the second spliced operation label and the target significance operation label to obtain a third derivative operation log queue for completing derivative analysis.

By means of the design, the second splicing processing is performed on the first spliced running label of the last photovoltaic power generation environment theme to obtain a second spliced running label, and the second spliced running label and the target significance running label are combined to obtain a third derivative running log queue with derivative analysis completed, so that the precision of the third derivative running log queue with derivative analysis completed can be improved.

For some possible embodiments, the first spliced operational label includes a plurality of photovoltaic power generation environmental themes; the performing a linear regression operation on the first spliced running label to obtain a target significance running label comprises: and performing linear regression operation on the first spliced running label of the first photovoltaic power generation environment theme to obtain a target significance running label.

The method comprises the steps of performing linear regression operation on a first spliced running label of a first photovoltaic power generation environment theme to obtain a target significance running label, and combining the first spliced running label and the target significance running label to obtain a third derivative running log queue for completing derivative analysis, so that disturbance of an AI model can be avoided as much as possible, and the precision of log derivative analysis and running state prediction is further improved.

For some possible embodiments, the performing significance running label mining on the first derived running journal queue to obtain a first significance running label pointed to by the first derived running journal queue comprises: performing significance operation label mining on a plurality of photovoltaic power generation environment themes on the first derivative operation log queue to obtain a plurality of photovoltaic power generation environment theme first significance operation labels pointed by the first derivative operation log queue; the performing significance running label mining on the second derived running log queue to obtain a second significance running label pointed by the second derived running log queue comprises: performing a plurality of photovoltaic power generation environment theme significance operation label mining on the second derivative operation log queue to obtain a plurality of photovoltaic power generation environment theme second significance operation labels pointed by the second derivative operation log queue; the performing a first stitching process in conjunction with the first salient running label and the second salient running label to obtain a first stitched running label comprises: for one photovoltaic power generation environment theme in a plurality of photovoltaic power generation environment themes, performing significance operation label splicing according to a first significance operation label of the photovoltaic power generation environment theme and a second significance operation label of the photovoltaic power generation environment theme to obtain a first spliced operation label of the photovoltaic power generation environment theme.

By the design, the first significant operation labels of the photovoltaic power generation environment themes pointed by the first derived operation log queue can be obtained by mining the significant operation labels of the photovoltaic power generation environment themes on the first derived operation log queue; and mining the plurality of photovoltaic power generation environment theme significance operation labels of the second derivative operation log queue, so as to obtain the second significance operation labels of the plurality of photovoltaic power generation environment themes pointed by the second derivative operation log queue. Therefore, by performing the mining of the plurality of photovoltaic power generation environment theme significance running labels on the first derivative running log queue and the second derivative running log queue, derivative running log information as complete as possible can be obtained. And executing first splicing treatment by utilizing the first significance operation labels of the plurality of photovoltaic power generation environment themes and the second significance operation labels of the plurality of photovoltaic power generation environment themes, so as to obtain first spliced operation labels of the plurality of photovoltaic power generation environment themes. According to the first spliced operation labels of the photovoltaic power generation environment themes, a more accurate and complete third derivative operation log queue can be obtained through derivative analysis.

For some possible embodiments, the performing significance running label splicing according to the first significance running label of the photovoltaic power generation environment theme and the second significance running label of the photovoltaic power generation environment theme to obtain a first spliced running label of the photovoltaic power generation environment theme includes: based on the fact that the photovoltaic power generation environment theme does not match with the last photovoltaic power generation environment theme, performing significance operation label splicing on a first significance operation label of the photovoltaic power generation environment theme, a second significance operation label of the photovoltaic power generation environment theme and a first spliced operation label of a next photovoltaic power generation environment theme of the photovoltaic power generation environment theme to obtain a first spliced operation label of the photovoltaic power generation environment theme; and/or performing significance operation label splicing on the first significance operation label of the photovoltaic power generation environment theme and the second significance operation label of the photovoltaic power generation environment theme based on the fact that the photovoltaic power generation environment theme is matched with the last photovoltaic power generation environment theme so as to obtain a first spliced operation label of the photovoltaic power generation environment theme.

By the design, the significance running labels of the photovoltaic power generation environment themes can be spliced as comprehensively as possible, so that the first spliced running labels which are as abundant and complete as possible are obtained.

According to another aspect of the present invention, there is also provided an information analysis system, including a processing engine, a network module, and a memory, the processing engine and the memory being in communication via the network module, the processing engine being configured to read a computer program from the memory and to execute the computer program, so as to implement the method described above.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.

The methods, systems, and/or programs of the figures will be further described in accordance with the exemplary embodiments. These exemplary embodiments will be described in detail with reference to the drawings. These exemplary embodiments are non-limiting exemplary embodiments in which reference numerals represent similar mechanisms throughout the various views of the drawings.

Fig. 1 is a block diagram illustrating an application scenario of an exemplary information analysis method for an above-water photovoltaic system according to some embodiments of the present invention.

FIG. 2 is a schematic diagram illustrating hardware and software components of an exemplary information analysis system, according to some embodiments of the invention.

Fig. 3 is a flow chart of an exemplary method and/or process for information analysis for an above-water photovoltaic system, according to some embodiments of the present invention.

Detailed Description

In order to better understand the technical solutions of the present invention, the following detailed descriptions of the technical solutions of the present invention are provided with the accompanying drawings and the specific embodiments, and it should be understood that the specific features in the embodiments and the examples of the present invention are the detailed descriptions of the technical solutions of the present invention, and are not limitations of the technical solutions of the present invention, and the technical features in the embodiments and the examples of the present invention may be combined with each other without conflict.

In the following detailed description, numerous specific details are set forth by way of examples in order to provide a thorough understanding of the relevant guidance. It will be apparent, however, to one skilled in the art that the present invention may be practiced without these specific details. In other instances, well-known methods, procedures, systems, compositions, and/or circuits have been described at a relatively high-level, without detail, in order to avoid unnecessarily obscuring aspects of the invention.

These and other features, functions, methods of execution, and combination of functions and elements of related elements in the structure disclosed in the present application, and the economics of production may become more apparent upon consideration of the following description with reference to the accompanying drawings, all of which form a part of this disclosure. It is to be expressly understood, however, that the drawings are for the purpose of illustration and description only and are not intended as a definition of the limits of the invention. It should be understood that the drawings are not to scale. It is to be expressly understood, however, that the drawings are for the purpose of illustration and description only and are not intended as a definition of the limits of the invention. It should be understood that the drawings are not to scale.

The present invention uses flow charts to illustrate the execution processes performed by a system according to an embodiment of the present invention. It should be expressly understood that the processes performed by the flowcharts may be performed out of order. Rather, these implementations may be performed in the reverse order or simultaneously. In addition, at least one other implementation may be added to the flowchart. One or more implementations may be deleted from the flowchart.

Fig. 1 is a block diagram illustrating an exemplary application environment 300 for an information analysis method for a marine photovoltaic system, which application environment 300 may include an information analysis system 100 and a marine photovoltaic system 200, according to some embodiments of the present invention.

In some embodiments, as shown in fig. 2, the information analysis system 100 may include a processing engine 110, a network module 120, and a memory 130, the processing engine 110 and the memory 130 communicating through the network module 120.

Processing engine 110 may process the relevant information and/or data to perform one or more of the functions described in this disclosure. For example, in some embodiments, processing engine 110 may include at least one processing engine (e.g., a single core processing engine or a multi-core processor). By way of example only, the Processing engine 110 may include a Central Processing Unit (CPU), an Application-Specific Integrated Circuit (ASIC), an Application-Specific Instruction Set Processor (ASIP), a Graphics Processing Unit (GPU), a Physical Processing Unit (PPU), a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), a Programmable Logic Device (PLD), a controller, a microcontroller Unit, a Reduced Instruction Set Computer (RISC), a microprocessor, or the like, or any combination thereof.

Network module 120 may facilitate the exchange of information and/or data. In some embodiments, the network module 120 may be any type or combination of wired or wireless network. Merely by way of example, the Network module 120 may include a cable Network, a wired Network, a fiber optic Network, a telecommunications Network, an intranet, the internet, a Local Area Network (LAN), a Wide Area Network (WAN), a Wireless Local Area Network (WLAN), a Metropolitan Area Network (MAN), a Public Switched Telephone Network (PSTN), a bluetooth Network, a Wireless personal Area Network, a Near Field Communication (NFC) Network, and the like, or any combination thereof. In some embodiments, the network module 120 may include at least one network access point. For example, the network module 120 may include wired or wireless network access points, such as base stations and/or network access points.

The Memory 130 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like. The memory 130 is used for storing a program, and the processing engine 110 executes the program after receiving the execution instruction.

It will be appreciated that the configuration shown in FIG. 2 is merely illustrative and that information analysis system 100 may include more or fewer components than shown in FIG. 2 or have a different configuration than shown in FIG. 2. The components shown in fig. 2 may be implemented in hardware, software, or a combination thereof.

Fig. 3 is a flow chart illustrating an exemplary method and/or process for information analysis for an above-water photovoltaic system, which is applied to the information analysis system 100 in fig. 1, according to some embodiments of the present invention, and further may include the technical solutions described in the following.

Step 31, the information analysis system performs downsampling operation on the received first above-water photovoltaic operation log queue to obtain a first downsampling description information queue pointed by the first above-water photovoltaic operation log queue.

For example, the information analysis system communicates with the photovoltaic system on water for obtaining a first photovoltaic operation log queue uploaded by the photovoltaic system on water. The photovoltaic operation log on water that includes in the first photovoltaic operation log queue on water is used for recording contents such as operating condition of photovoltaic system on water. The down-sampling operation can be understood as performing compression processing or reduction processing on the first overwater photovoltaic operation log queue, the obtained first down-sampling description information queue can be understood as a sequence formed by a plurality of down-sampling description information, and the down-sampling description information can be recorded in a matrix or list form.

In some possible examples, the overwater photovoltaic operation log records included in the first overwater photovoltaic operation log queue may be real-time operation states of the overwater photovoltaic system, such as stress analysis conditions between structures/architectures of the overwater photovoltaic system, photoelectric energy conversion loss, water displacement change and other operation states. The stress analysis conditions between the structures/architectures of the above-water photovoltaic system can include torsion, stress variation conditions between connecting beams, buoyancy devices and the like. Photoelectric energy conversion loss can represent photoelectric conversion efficiency, and displacement change can represent the equilibrium of the whole photovoltaic system on water. Of course, the overwater photovoltaic operation log may further include other types of operation states, and a person skilled in the art may flexibly select different operation states for analysis according to actual conditions, which are not listed herein.

In other possible examples, the above-water photovoltaic operation log may be recorded in a text form, a graphic form, or a text + image form, which is not limited herein.

Further, the downsampling operation is used for performing key state feature extraction on the first overwater photovoltaic operation log queue to obtain downsampling description information, the downsampling description information can record the operation state of the overwater photovoltaic system in the form of feature vectors, for example, the downsampling description information Vec1 is used for recording stress analysis conditions between structures/architectures of the overwater photovoltaic system, the downsampling description information Vec2 is used for recording photoelectric energy conversion loss of the overwater photovoltaic system, the downsampling description information Vec2 is used for recording displacement changes of the overwater photovoltaic system, and Vec1, Vec2 and Vec3 can be feature matrices or feature lists.

Through downsampling operation/feature mining processing, not only can key state feature extraction be realized, but also processing load caused by redundant information (such as some fixed and default field contents) in the overwater photovoltaic operation log can be reduced, and therefore the efficiency of subsequent state prediction is improved.

In some possible embodiments, the down-sampling operation performed on the received first photovoltaic operation log queue on water in step 31 to obtain a first down-sampling description information queue pointed by the first photovoltaic operation log queue on water may be implemented by the technical solutions described in step 311 and step 312.

Step 311, performing set significance operation label mapping on the received first underwater photovoltaic operation log queue to obtain a first simplified description information queue pointed by the first underwater photovoltaic operation log queue.

For example, the mapping of the set significance operation label can be understood as cosine mapping, and the simplified description information queue can be understood as sparse description information queue.

And step 312, performing unconstrained updating on the first simplified description information queue to obtain a first downsampling description information queue pointed by the first above-water photovoltaic operation log queue.

By way of example, a non-constrained update may be understood as a random mapping.

It can be understood that, through steps 311 and 312, the design is such that the feature size is reduced by performing the set significance operation label mapping and the unconstrained updating on the first overwater photovoltaic operation log queue to obtain a first downsampling descriptor queue with a smaller feature size for the subsequent derivative analysis processing. By performing the set significance operation label mapping and the unconstrained updating on the first overwater photovoltaic operation log queue to perform the downsampling operation, compared with the related art, the operation difficulty can be reduced. By performing the set significance operation label mapping and the unconstrained updating on the first overwater photovoltaic operation log queue to perform the downsampling operation, the processing timeliness can be improved.

And 32, combining the first downsampling description information queue with the information analysis system to perform derivative analysis so as to obtain a second downsampling description information queue for finishing derivative analysis.

By way of example, a derivative analysis may be understood as a prospective analysis or a predictive analysis, such as a second downsampled descriptor queue determined a time later by the first downsampled descriptor queue. It is to be understood that the second downsampling descriptor queue is subsequent to the first downsampling descriptor queue, and the second downsampling descriptor queue may be considered as a predicted operating condition characteristic of the above-water photovoltaic system.

For example, the predicted characteristic corresponding to Vec1 is Vec1-0, the predicted characteristic corresponding to Vec2 is Vec2-0, and the predicted characteristic corresponding to Vec3 is Vec 3-0. If the characteristic field of Vec1 is "buoyancy device-011" (characterizing an internal pressure anomaly of the buoyancy device), then the characteristic field of Vec1-0 is "buoyancy device-010" (characterizing the presence of an air leak and a decrease in buoyancy of the buoyancy device). If the characteristic field of Vec2 is "photoelectric conversion rate-001" (indicating that the increase of the photoelectric conversion rate is reduced), the characteristic field of Vec2-0 is "photoelectric conversion rate-002" (indicating that there is a tendency toward stabilization of the photoelectric conversion rate). If the characteristic field of Vec3 is "Change in Displacement-100" (characteristic of abnormal Displacement changes), then the characteristic field of Vec3-0 is "Change in Displacement-101" (characteristic of abnormal Displacement changes that may cause abnormal fluctuations in the marine photovoltaic system).

For some possible embodiments, performing the derivative analysis in conjunction with the first downsampling descriptor queue to obtain a second downsampling descriptor queue for which the derivative analysis is completed as described in step 32 may include the following: and loading the first downsampling description information queue to a first staged AI model, and obtaining a second downsampling description information queue pointed by the first downsampling description information queue by utilizing the first staged AI model for derivative analysis. For example, the staged AI model may be understood as a sub-neural network. By means of the design, the first staged AI model is combined with the first downsampling description information queue to perform derivative analysis so as to obtain a second downsampling description information queue for completing derivative analysis, and therefore derivative analysis precision and derivative analysis timeliness can be improved.

And step 33, the information analysis system performs log reduction operation on the second downsampling description information queue to obtain a first derivative running log queue pointed by the second downsampling description information queue.

For example, a log reduction operation may be understood as describing an information translation process or a feature encoding process, such as reducing a feature vector to a corresponding information content, such as reducing a second downsampled descriptor queue to a first derived run log queue. The first derived operation log queue comprises a plurality of first derived operation logs, and the first derived operation logs are used for recording the possible later operation state of the overwater photovoltaic system, so that a decision basis is provided for the subsequent failure prediction analysis processing.

It can be understood that the core content of the second downsampling description information queue is determined based on the first downsampling description information queue and is obtained through feature prediction and feature derivation, so that the complete running log can be prevented from being directly subjected to prediction processing, the operation cost is obviously reduced, and the state prediction efficiency is improved.

By taking Vec1-0, Vec2-0 and Vec3-0 as examples, after feature restoration/feature translation is performed on Vec1-0, Vec2-0 and Vec3-0, the prediction logs included in the first derived run log queue obtained can be, but are not limited to, the following three types.

Prediction log 1: an anomaly in the internal pressure of the buoyancy device may cause the buoyancy device to leak and the buoyancy to decrease.

Prediction log 2: the photoelectric conversion rate tends to be stable due to the decrease in the increase amount of the photoelectric conversion rate.

Prediction log 3: abnormal fluctuation of the overwater photovoltaic system can be caused due to abnormal water discharge variation.

Through the analysis of the prediction log, the (abnormal) operation state which may occur after the photovoltaic system on the water is buoyancy reduction of the buoyancy equipment, severe fluctuation of the system on the water and the like, and based on the (abnormal) operation state, related maintenance work (such as replacement/repair of the buoyancy equipment in advance, inspection of a drainage system in advance and the like) can be guided in advance.

It can be understood that, in the embodiment of the present invention, artificial intelligence technology (such as related neural network models) may be used to perform processing such as feature mining, feature prediction, and feature restoration, and on the premise of the technical solution provided in the embodiment of the present invention, a person skilled in the art may select a corresponding neural network model according to actual requirements to apply, including but not limited to CNN, GCN, LSTM, and the like, and the architecture of the neural network model may also be adjusted according to actual requirements, which is not limited herein.

On the basis of the above, the performing, by step 33, a log reduction operation on the second downsampling descriptor queue to obtain a first derived running log queue pointed to by the second downsampling descriptor queue may include the technical solutions described in step 331 and step 332.

Step 331, performing a log reduction operation on the second downsampled descriptor queue to obtain a second simplified descriptor queue pointed by the second downsampled descriptor queue.

In some possible embodiments, the performing a log reduction operation on the second downsampled descriptor queue to obtain a second simplified descriptor queue pointed to by the second downsampled descriptor queue as described in step 331 includes: and executing cyclic processing on the second downsampling description information queue by adopting a trigger algorithm of a dynamic judgment value to obtain a second simplified description information queue pointed by the second downsampling description information queue.

For example, the triggering algorithm of the dynamic decision value may be understood as an activation function Relu with an adjustable threshold. The loop processing can be understood as iteration operation, and by adopting the design, the loop processing is performed on the second downsampling description information queue by adopting a trigger algorithm of a dynamic judgment value, so that a second simplified description information queue pointed by the second downsampling description information queue can be obtained in time, and the timeliness of running log derivative analysis can be improved.

Step 332, performing significance running label restoration on the second simplified description information queue to obtain a first derived running log queue pointed by the second downsampling description information queue.

It can be understood that, when applied to steps 331 and 332, the design is such that a second simplified description information queue is obtained by performing a log reduction operation on the second downsampling description information queue, and a set significance operation tag reduction is performed on the second simplified description information queue, each downsampling description information in the second downsampling description information queue can be respectively translated into a derivative operation log which is consistent with a derivative operation log information distribution structure in the first overwater photovoltaic operation log queue.

In some optional embodiments, after obtaining the first derived running log queue pointed to by the second downsampled description information queue described in step 33, the method may further include step 34 and step 35.

And step 34, performing derivative analysis by combining at least part of the photovoltaic operation logs on the water in the first overwater photovoltaic operation log queue to obtain a second derivative operation log queue for finishing derivative analysis.

In some possible embodiments, performing a derivative analysis in conjunction with at least some of the water photovoltaic operational logs in the first queue of water photovoltaic operational logs as described in step 34 to obtain a second queue of derivative operational logs that complete the derivative analysis includes: and loading at least part of the water photovoltaic operation logs in the first water photovoltaic operation log queue to a second stage AI model, and performing derivative analysis by using the second stage AI model to obtain a second derivative operation log queue pointed by the at least part of the water photovoltaic operation logs.

For example, the at least part of the water photovoltaic operation logs include X number of water photovoltaic operation logs currently received in the first water photovoltaic operation log queue, where X is a positive integer, and the number of the water photovoltaic operation logs in the first water photovoltaic operation log queue is not less than X.

And step 35, combining the first derived running log queue and the second derived running log queue to obtain a third derived running log queue for completing derived analysis.

In some other embodiments, the combining the first derived logbook queue and the second derived logbook queue to obtain a third derived logbook queue for performing the derived analysis, which is described in step 35, may include the technical solutions described in steps 351 to 354.

Step 351, performing significance running label mining on the first derived running log queue to obtain a first significance running label pointed by the first derived running log queue.

For some possible embodiments, performing saliency oplabel mining on the first derived oplog queue, as described in step 351, to obtain a first saliency oplabel pointed to by the first derived oplog queue, may include: and performing a plurality of photovoltaic power generation environment theme significance operation label mining on the first derivative operation log queue to obtain a plurality of photovoltaic power generation environment theme first significance operation labels pointed by the first derivative operation log queue.

And 352, performing significance running label mining on the second derived running log queue to obtain a second significance running label pointed by the second derived running log queue.

For example, significant operation label mining may be understood as feature extraction, where significant operation labels correspond to operational features of an above-water photovoltaic system.

Further, performing significance run label mining on the second derived run log queue to obtain a second significance run label pointed to by the second derived run log queue as described in step 352, comprising: and performing multiple photovoltaic power generation environment theme significance operation label mining on the second derivative operation log queue to obtain multiple photovoltaic power generation environment theme second significance operation labels pointed by the second derivative operation log queue.

And 353, combining the first significance running label and the second significance running label to execute a first splicing process so as to obtain a first spliced running label.

By way of example, the stitching process may be understood as a fusion process.

For some possible embodiments, performing a first stitching process in conjunction with the first salient running label and the second salient running label to obtain a first stitched running label as described in step 353 includes: for one photovoltaic power generation environment theme in a plurality of photovoltaic power generation environment themes, performing significance operation label splicing according to a first significance operation label of the photovoltaic power generation environment theme and a second significance operation label of the photovoltaic power generation environment theme to obtain a first spliced operation label of the photovoltaic power generation environment theme.

In some possible embodiments, the performing of the saliency operating label stitching according to the first saliency operating label of the photovoltaic power generation environment theme and the second saliency operating label of the photovoltaic power generation environment theme described in the above steps to obtain the first stitched operating label of the photovoltaic power generation environment theme includes at least one of the following embodiments a and B.

Embodiment a, based on that the photovoltaic power generation environment theme does not match the last photovoltaic power generation environment theme, perform saliency operation label stitching on the first saliency operation label of the photovoltaic power generation environment theme, the second saliency operation label of the photovoltaic power generation environment theme, and the first stitched operation label of the next photovoltaic power generation environment theme of the photovoltaic power generation environment theme to obtain the first stitched operation label of the photovoltaic power generation environment theme.

Embodiment B, based on the photovoltaic power generation environment theme being matched to the last photovoltaic power generation environment theme, performing significance operation label stitching on the first significance operation label of the photovoltaic power generation environment theme and the second significance operation label of the photovoltaic power generation environment theme to obtain a first stitched operation label of the photovoltaic power generation environment theme.

When the method is applied to the embodiment A and/or the embodiment B, the significance running labels of all the photovoltaic power generation environment topics can be spliced as comprehensively as possible, so that the first spliced running labels which are as abundant and complete as possible are obtained.

It can be understood that, when applied to steps 351 to 353, by performing the mining of the plurality of photovoltaic power generation environment theme significance running labels on the first derivative running log queue, the first significance running label of the plurality of photovoltaic power generation environment themes pointed by the first derivative running log queue can be obtained; by performing the significance operation label mining on the plurality of photovoltaic power generation environment themes on the second derivative operation log queue, second significance operation labels of the plurality of photovoltaic power generation environment themes pointed by the second derivative operation log queue can be obtained. Therefore, by performing the mining of the plurality of photovoltaic power generation environment theme significance running labels on the first derived running log queue and the second derived running log queue, the derived running log information as complete as possible can be obtained. And executing first splicing treatment by utilizing the first significance running labels of the plurality of photovoltaic power generation environment themes and the second significance running labels of the plurality of photovoltaic power generation environment themes, so as to obtain first spliced running labels of the plurality of photovoltaic power generation environment themes. According to the first spliced running labels of the multiple photovoltaic power generation environment themes, a more accurate and complete third derivative running log queue can be obtained through derivative analysis.

And 354, combining the first spliced running label to obtain a third derivative running log queue for finishing derivative analysis.

In some possible embodiments, the step 354 of combining the first spliced running label to obtain a third derived running log queue for performing derived analysis may include the steps 3541 and 3542.

Step 3541, a linear regression operation is performed on the first spliced running label to obtain a target significance running label.

For example, a linear regression operation may be understood as a residual process.

For some possible implementations, the first spliced operational label includes a plurality of photovoltaic power generation environment themes. Based thereon, performing a linear regression operation on the first stitched running label to obtain a target saliency running label as described in step 3541, comprising: and performing linear regression operation on the first spliced running label of the first photovoltaic power generation environment theme to obtain a target significance running label. In this way, a target significance running label is obtained by performing linear regression operation on the first spliced running label of the first photovoltaic power generation environment theme, and a third derivative running log queue for completing derivative analysis is obtained by combining the first spliced running label and the target significance running label, so that disturbance of the AI model can be avoided as much as possible, and the precision of log derivative analysis and running state prediction is further improved.

Step 3542, combining the first stitched run tag and the target significance run tag to obtain a third derived run log queue that completes the derived analysis.

In some possible examples, the first spliced operational signature includes a plurality of photovoltaic power generation environmental themes. Each photovoltaic power generation environment topic corresponds to a different environmental quality level, and based on this, combining the first stitched operational label and the target significance operational label to obtain a third derivative operational log queue for completing the derivative analysis as described in step 3542 includes: performing second splicing processing on the first spliced running label of the last photovoltaic power generation environment theme to obtain a second spliced running label; and combining the second spliced running label and the target significance running label to obtain a third derivative running log queue for completing derivative analysis. By means of the design, the second splicing processing is performed on the first spliced running label of the last photovoltaic power generation environment theme to obtain a second spliced running label, and the second spliced running label and the target significance running label are combined to obtain a third derivative running log queue with derivative analysis completed, so that the precision of the third derivative running log queue with derivative analysis completed can be improved.

The method is applied to steps 3541 and 3542, linear regression operation is performed on the first spliced running label to obtain a target significance running label, and a third derivative running log queue for completing derivative analysis is obtained by combining the first spliced running label and the target significance running label, so that disturbance of an AI model can be avoided as much as possible, and the precision of log derivative analysis and running state prediction is further improved. Further, aiming at the debugging process of the AI model, by executing the linear regression operation, the convergence of the AI model can be realized as fast as possible so as to ensure that the AI model quickly reaches a stable state.

Applied to steps 351-354, by combining the first significant running label extracted from the first derived running log queue and the second significant running label extracted from the second derived running log queue, running log derived analysis can be executed based on the derived running log information as complete as possible, so that the precision of the log derived analysis and the running state prediction can be improved.

It can be understood that, through steps 34 and 35, by performing a derivative analysis by combining at least some of the photovoltaic operation logs in the first waterborne photovoltaic operation log queue to obtain a second derivative operation log queue in which the derivative analysis is completed, and combining the first derivative operation log queue and the second derivative operation log queue to obtain a third derivative operation log queue in which the derivative analysis is completed, the operation log derivative analysis can be performed based on derivative operation log information as complete as possible, so that the accuracy of the derivative analysis result can be further improved.

In some alternative embodiments, after performing a log reduction operation on the second downsampled description information queue to obtain a first derived running log queue pointed to by the second downsampled description information queue, the method further comprises: acquiring a state early warning message set aiming at a first derived running log queue, wherein the state early warning message set comprises at least two state early warning messages; obtaining the importance degree between each state early warning message in the state early warning message set and the first derived running log queue; sorting the state early warning messages according to the corresponding importance degrees of the state early warning messages and the early warning categories of the state early warning messages to obtain corresponding state early warning message sequences; generating a target operation state adjustment indication sequence aiming at the first derived operation log queue based on the state early warning message sequence, wherein the target operation state adjustment indication sequence comprises at least two target operation state adjustment indications. By the design, the subsequent target running state adjustment indication sequence can be completely and accurately obtained.

In some alternative embodiments, the sorting the state early warning messages according to the importance degrees corresponding to the state early warning messages and the early warning categories of the state early warning messages to obtain corresponding state early warning message sequences specifically includes: according to the corresponding importance degree of each state early warning message and the early warning category of each state early warning message, disassembling each state early warning message to obtain at least two state early warning message subsets; and sorting each state early warning message subset, and sorting each state early warning message in each state early warning message subset respectively to obtain the state early warning message sequence. By the design, the integrity of the state early warning message sequence can be guaranteed.

In some alternative embodiments, the disassembling, according to the importance degree corresponding to each state warning message and the warning category of each state warning message, each state warning message to obtain at least two state warning message subsets specifically includes: respectively carrying out global integration on the early warning categories of the state early warning messages according to the corresponding importance degrees of the state early warning messages to obtain the attention early warning categories of the state early warning messages; and grouping the state early warning messages according to the attention early warning categories of the state early warning messages to obtain at least two state early warning message subsets. By the design, the discrimination between the state early warning message subsets can be guaranteed.

In some alternative embodiments, the sorting among the subsets of state early warning messages, and the sorting of each state early warning message in each subset of state early warning messages, respectively, to obtain the sequence of state early warning messages specifically includes: sorting each state early warning message subset according to the number of state early warning messages contained in each state early warning message subset; and for each state early warning message subset, respectively executing the following operations: sorting each state early warning message in the state early warning message subset according to the association degree between the early warning type of each state early warning message in the state early warning message subset and the state early warning message subset; and generating the state early warning message sequence based on the sorting result among the state early warning message subsets and the sorting result of each state early warning message in each state early warning message subset. By the design, the integrity of the state early warning message sequence can be guaranteed, and the loss of part of the state early warning messages is avoided.

The skilled person can unambiguously determine some preset, reference, predetermined, set and target technical features/terms, such as threshold values, threshold intervals, threshold ranges, etc., from the above disclosure. For some technical characteristic terms which are not explained, the technical solution can be clearly and completely implemented by those skilled in the art by reasonably and unambiguously deriving the technical solution based on the logical relations in the previous and following paragraphs. Prefixes of unexplained technical feature terms, such as "first", "second", "previous", "next", "current", "history", "latest", "best", "target", "specified", and "real-time", etc., can be unambiguously derived and determined from the context. Suffixes of technical feature terms not to be explained, such as "list", "feature", "sequence", "set", "matrix", "unit", "element", "track", and "list", etc., can also be derived and determined unambiguously from the foregoing and the following.

The foregoing disclosure of embodiments of the present invention will be apparent to those skilled in the art. It should be understood that the process of deriving and analyzing technical terms, which are not explained, by those skilled in the art based on the above disclosure is based on the contents described in the present invention, and thus the above contents are not an inventive judgment of the overall scheme.

Having thus described the basic concept, it will be apparent to those skilled in the art that the foregoing detailed disclosure is to be considered as illustrative and not restrictive. Various modifications, adaptations, and alternatives may occur to one skilled in the art, though not explicitly described herein. Such modifications, improvements and adaptations are proposed within the present invention and are intended to be within the spirit and scope of the exemplary embodiments of the present invention.

Meanwhile, the present invention uses specific terms to describe embodiments of the present invention. Such as "one embodiment," "an embodiment," and/or "some embodiments" means a feature, structure, or characteristic described in connection with at least one embodiment of the invention. Therefore, it is emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various portions of this specification are not necessarily all referring to the same embodiment. Furthermore, certain features, structures, or characteristics of at least one embodiment of the invention may be combined as suitable.

In addition, those skilled in the art will recognize that the various aspects of the invention may be illustrated and described in terms of several patentable species or situations, including any new and useful combination of processes, machines, articles of manufacture, or materials, or any new and useful modifications thereto. Accordingly, aspects of the present invention may be embodied entirely in hardware, entirely in software (including firmware, resident software, micro-code, etc.) or in a combination of hardware and software. The above hardware or software may be referred to as a "unit", "component", or "system". Furthermore, aspects of the present invention may be embodied as a computer product, located in at least one computer-readable medium, comprising computer-readable program code.

A computer readable signal medium may comprise a propagated data signal with computer program code embodied therein, for example, on a baseband or as part of a carrier wave. The propagated signal may take any of a variety of forms, including electromagnetic, optical, and the like, or any suitable combination. A computer readable signal medium may be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code on a computer readable signal medium may be propagated over any suitable medium, including radio, electrical cable, fiber optic cable, RF, or the like, or any combination of the preceding.

Computer program code required for the execution of aspects of the present invention may be written in any combination of one or more programming languages, including object oriented programming, such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C + +, C #, VB.NET, Python, and the like, or similar conventional programming languages, such as the "C" programming language, Visual Basic, Fortran 2003, Perl, COBOL 2002, PHP, ABAP, dynamic programming languages, such as Python, Ruby, and Groovy, or other programming languages. The programming code may execute entirely on the user's computer, as a stand-alone software package, partly on the user's computer, partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any network format, such as a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet), or in a cloud computing environment, or as a service, such as a software as a service (SaaS).

Additionally, unless explicitly stated in the claims, the order of processing elements and sequences, use of numerical letters, or use of other designations in the present disclosure is not intended to limit the order of the processes and methods of the present disclosure. While various presently contemplated embodiments of the invention have been discussed in the foregoing disclosure by way of example, it should be understood that such detail is solely for that purpose and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements that are within the spirit and scope of the embodiments of the invention. For example, although the system components described above may be implemented by hardware means, they may also be implemented by software-only solutions, such as installing the described system on an existing server or mobile device.

It should also be appreciated that in the foregoing description of embodiments of the invention, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of at least one embodiment of the invention. This method of disclosure, however, is not intended to suggest that more features are required than are set forth in the claims. Indeed, the embodiments may be characterized as having less than all of the features of a single disclosed embodiment.

Claims

1. An information analysis method for an overwater photovoltaic system is applied to the information analysis system, and the method at least comprises the following steps:

performing downsampling operation on the received first photovoltaic operation log queue on the water to obtain a first downsampling description information queue pointed by the first photovoltaic operation log queue on the water;

performing a derivative analysis in conjunction with the first downsampling descriptor queue to obtain a second downsampling descriptor queue that completes the derivative analysis;

performing log reduction operation on the second downsampling description information queue to obtain a first derivative running log queue pointed by the second downsampling description information queue;

wherein after the obtaining the first derived run log queue to which the second downsampled description information queue points, the method further comprises: performing derivative analysis by combining at least part of the photovoltaic operation logs on the water in the first overwater photovoltaic operation log queue to obtain a second derivative operation log queue for finishing derivative analysis; combining the first derived run journal queue and the second derived run journal queue to obtain a third derived run journal queue for which derived analysis is completed;

Wherein the performing a derivative analysis in conjunction with at least some of the waterborne photovoltaic execution logs in the first waterborne photovoltaic execution log queue to obtain a second derivative execution log queue for which the derivative analysis is completed comprises: loading at least part of the water photovoltaic operation logs in the first water photovoltaic operation log queue to a second stage AI model, and performing derivative analysis by using the second stage AI model to obtain a second derivative operation log queue pointed by the at least part of the water photovoltaic operation logs;

the at least part of the overwater photovoltaic operation logs comprises X overwater photovoltaic operation logs currently received in the first overwater photovoltaic operation log queue, wherein X is a positive integer, and the number of the overwater photovoltaic operation logs in the first overwater photovoltaic operation log queue is not less than X;

wherein the combining the first derived run journal queue and the second derived run journal queue to obtain a third derived run journal queue that completes a derived analysis comprises: performing significance running label mining on the first derivative running log queue to obtain a first significance running label pointed by the first derivative running log queue; performing significance running label mining on the second derived running log queue to obtain a second significance running label pointed by the second derived running log queue; performing a first stitching process in conjunction with the first salient running label and the second salient running label to obtain a first stitched running label; combining the first spliced running label to obtain a third derivative running log queue for finishing derivative analysis;

Wherein the combining the first spliced running label to obtain a third derived running log queue of completed derived analytics comprises: performing a linear regression operation on the first spliced running label to obtain a target significance running label; combining the first spliced running label and the target significance running label to obtain a third derivative running log queue for completing derivative analysis;

wherein the first spliced operational label comprises a plurality of photovoltaic power generation environmental themes; the combining the first spliced running label and the target saliency running label to obtain a third derived running log queue completing the derived analysis comprises: performing second splicing processing on the first spliced running label of the last photovoltaic power generation environment theme to obtain a second spliced running label; combining the second spliced running label and the target significance running label to obtain a third derivative running log queue for completing derivative analysis;

wherein the first spliced operational label comprises a plurality of photovoltaic power generation environmental themes; the performing a linear regression operation on the first spliced running label to obtain a target saliency running label comprises: performing linear regression operation on a first spliced running label of a first photovoltaic power generation environment theme to obtain a target significance running label;

Wherein the performing significance running label mining on the first derived running log queue to obtain a first significance running label pointed to by the first derived running log queue comprises: performing a plurality of photovoltaic power generation environment theme significance running label mining on the first derivative running log queue to obtain a plurality of photovoltaic power generation environment theme first significance running labels pointed by the first derivative running log queue;

the performing significance running label mining on the second derived running log queue to obtain a second significance running label pointed to by the second derived running log queue comprises: performing a plurality of photovoltaic power generation environment theme significance operation label mining on the second derivative operation log queue to obtain a plurality of photovoltaic power generation environment theme second significance operation labels pointed by the second derivative operation log queue;

the performing a first stitching process in conjunction with the first salient running label and the second salient running label to obtain a first stitched running label includes: for one photovoltaic power generation environment theme in a plurality of photovoltaic power generation environment themes, performing significance operation label splicing according to a first significance operation label of the photovoltaic power generation environment theme and a second significance operation label of the photovoltaic power generation environment theme to obtain a first spliced operation label of the photovoltaic power generation environment theme;

Wherein the performing significance running label splicing according to the first significance running label of the photovoltaic power generation environment theme and the second significance running label of the photovoltaic power generation environment theme to obtain a first spliced running label of the photovoltaic power generation environment theme comprises at least one of:

based on the fact that the photovoltaic power generation environment theme does not match the last photovoltaic power generation environment theme, performing significance operation label splicing on a first significance operation label of the photovoltaic power generation environment theme, a second significance operation label of the photovoltaic power generation environment theme and a first spliced operation label of a next photovoltaic power generation environment theme of the photovoltaic power generation environment theme to obtain a first spliced operation label of the photovoltaic power generation environment theme;

performing significance running label splicing on a first significance running label of the photovoltaic power generation environment theme and a second significance running label of the photovoltaic power generation environment theme based on the fact that the photovoltaic power generation environment theme is matched with the last photovoltaic power generation environment theme so as to obtain a first spliced running label of the photovoltaic power generation environment theme;

wherein after performing a log reduction operation on the second downsampled description information queue to obtain a first derived run log queue pointed to by the second downsampled description information queue, the method further comprises: acquiring a state early warning message set aiming at a first derived running log queue, wherein the state early warning message set comprises at least two state early warning messages; obtaining the importance degree between each state early warning message in the state early warning message set and the first derived running log queue; sorting the state early warning messages according to the corresponding importance degree of the state early warning messages and the early warning categories of the state early warning messages to obtain corresponding state early warning message sequences; generating a target running state adjustment indication sequence aiming at the first derived running log queue based on the state early warning message sequence, wherein the target running state adjustment indication sequence comprises at least two target running state adjustment indications;

The method for sorting the state early warning messages according to the importance degrees corresponding to the state early warning messages and the early warning categories of the state early warning messages to obtain the corresponding state early warning message sequences comprises the following steps: according to the corresponding importance degree of each state early warning message and the early warning category of each state early warning message, disassembling each state early warning message to obtain at least two state early warning message subsets; sorting all state early warning message subsets, and sorting all state early warning messages in all state early warning message subsets respectively to obtain a state early warning message sequence;

wherein, the disassembling each state early warning message according to the corresponding importance degree of each state early warning message and the early warning category of each state early warning message to obtain at least two state early warning message subsets comprises: respectively carrying out global integration on the early warning categories of the state early warning messages according to the corresponding importance degrees of the state early warning messages to obtain the attention early warning categories of the state early warning messages; grouping the state early warning messages according to the attention early warning categories of the state early warning messages to obtain at least two state early warning message subsets;

The sorting of the state early warning message subsets and the sorting of the state early warning messages in the state early warning message subsets to obtain the state early warning message sequence includes: sorting each state early warning message subset according to the number of state early warning messages contained in each state early warning message subset; and for each state early warning message subset, respectively executing the following operations: sorting each state early warning message in the state early warning message subset according to the association degree between the early warning type of each state early warning message in the state early warning message subset and the state early warning message subset; generating the state early warning message sequence based on the sorting result among the state early warning message subsets and the sorting result of each state early warning message in each state early warning message subset;

wherein:

the down-sampling operation is key state feature extraction;

the down-sampling description information queue is a feature vector;

the derivative analysis is predictive analysis;

the significant operation label mining is feature mining;

The salient operation labels are operation features.

2. The method of claim 1, wherein the performing a downsampling operation on the received first overwater photovoltaic running log queue to obtain a first downsampled description information queue pointed to by the first overwater photovoltaic running log queue comprises:

performing set significance operation label mapping on a received first photovoltaic operation log queue on water to obtain a first simplified description information queue pointed by the first photovoltaic operation log queue on water;

performing non-constrained updating on the first simplified description information queue to obtain a first down-sampling description information queue pointed by the first above-water photovoltaic operation log queue;

wherein:

the set significance operation label is mapped to a cosine map.

3. The method according to any one of claims 1-2, wherein the performing the derivative analysis in combination with the first downsampling descriptor queue to obtain a second downsampling descriptor queue for performing the derivative analysis comprises:

and loading the first downsampling description information queue to a first staged AI model, and obtaining a second downsampling description information queue pointed by the first downsampling description information queue by utilizing the first staged AI model for derivative analysis.

4. The method of claim 2, wherein performing a log reduction operation on the second downsampled descriptor queue to obtain a first derived running log queue pointed to by the second downsampled descriptor queue comprises:

performing log reduction operation on the second downsampling description information queue to obtain a second simplified description information queue pointed by the second downsampling description information queue;

and performing significance operation label restoration on the second simplified description information queue to obtain a first derivative operation log queue pointed by the second downsampling description information queue.

5. The method of claim 4, wherein performing a log reduction operation on the second downsampled descriptor queue to obtain a second simplified descriptor queue pointed to by the second downsampled descriptor queue comprises:

executing cyclic processing on the second downsampling description information queue by adopting a trigger algorithm of a dynamic judgment value to obtain a second simplified description information queue pointed by the second downsampling description information queue;

wherein:

the triggering algorithm of the dynamic judgment value is an activation function with adjustable threshold value.

6. An information analysis system comprising a processing engine, a network module and a memory, the processing engine and the memory being in communication via the network module, the processing engine being configured to read a computer program from the memory and to execute to implement the method of any one of claims 1 to 5.