Multifunctional monitoring method and system for solar power generation system

Abstract

According to the multifunctional monitoring method and system of the solar power generation system provided by the invention, the target multifunctional monitoring data needing to be processed and the important knowledge field information of the target needing to be processed are fused; the hidden knowledge field information of the target needing to be processed and the important knowledge field information of the target needing to be processed are synthesized, more knowledge field information of the target needing to be processed is extracted, and multifunctional monitoring data can be accurately determined. Therefore, the monitoring accuracy and reliability of the solar power generation system can be improved. Besides, through the target multifunctional monitoring data needing to be processed, the target needing to be processed is monitored, a solar power generation monitoring result is obtained, the monitoring accuracy is guaranteed, normal work of equipment can be guaranteed, and therefore the stability of the power generation amount is guaranteed.

Description

Technical Field

[0001] The present application relates to the technical field of data multi-functional monitoring, and more specifically, to a multi-functional monitoring method and system for a solar power generation system. Background Art

[0002] A solar power generation system is composed of a solar cell array, a solar controller, and a battery (bank). If the solar power generation system is to output a power supply of AC 220V or 110V, an inverter also needs to be configured.

[0003] Currently, it is very difficult to repair a solar power generation system because the area where the solar power generation system is laid is particularly large or the laying location is very remote. Therefore, when a fault occurs in the solar power generation system, each point needs to be checked to determine the location of the fault. Therefore, there is an urgent need for a multi-functional monitoring method for a solar power generation system to overcome the above problems. Summary of the Invention

[0004] To improve the technical problems existing in the related art, the present application provides a multi-functional monitoring method and system for a solar power generation system.

[0005] In a first aspect, a multi-functional monitoring method for a solar power generation system is provided, including: through an energy conversion level prediction model of a solar power generation efficiency gradient, performing knowledge field extraction processing of the solar power generation efficiency gradient on the power generation monitoring data of solar devices to be processed, obtaining knowledge field information of a target corresponding to the solar power generation efficiency gradient to be processed as hidden knowledge field information of the target to be processed; performing prediction processing in combination with the hidden knowledge field information of the target to be processed to obtain multi-functional monitoring data of the target to be processed; performing important knowledge field extraction processing on the power generation monitoring data of the solar devices to be processed to obtain important knowledge field information of the target to be processed; performing fusion processing on the multi-functional monitoring data of the target to be processed in the power generation monitoring data of the solar devices to be processed and the important knowledge field information of the target to be processed to obtain a fusion result of the target to be processed; and performing monitoring processing in combination with the fusion result of the target to be processed to obtain a solar power generation monitoring result.

[0006] In this application, the prediction model for predicting the power generation monitoring data of a solar energy device includes multiple energy conversion level prediction models, and the multiple energy conversion level prediction models respectively correspond to different solar power generation efficiency gradients; the prediction process is performed by combining the hidden knowledge field information of the target to be processed to obtain the target multi-functional monitoring data to be processed, including: through the energy conversion level prediction model of the solar power generation efficiency gradient, and combining the knowledge field information of the target to be processed corresponding to the solar power generation efficiency gradient for hidden prediction processing to obtain the target multi-functional monitoring data of the target to be processed corresponding to the solar power generation efficiency gradient; the target multi-functional monitoring data of the target to be processed corresponding to multiple solar power generation efficiency gradients is fused to obtain the target multi-functional monitoring data to be processed.

[0007] In this application, the hidden prediction process is performed by combining the knowledge field information of the target to be processed corresponding to the solar power generation efficiency gradient to obtain the target multi-functional monitoring data of the target to be processed corresponding to the solar power generation efficiency gradient, including: performing knowledge field deletion processing on the knowledge field information of the target to be processed corresponding to the solar power generation efficiency gradient to obtain the deleted knowledge field information; performing weighted processing on the deleted knowledge field information to obtain the target multi-functional monitoring data of the target to be processed corresponding to the solar power generation efficiency gradient.

[0008] In this application, the fusion process of the target multi-functional monitoring data of the target to be processed corresponding to multiple solar power generation efficiency gradients to obtain the target multi-functional monitoring data to be processed includes: when there are target multi-functional monitoring data corresponding to two different solar power generation efficiency gradients, the difference between the target multi-functional monitoring data of the two different solar power generation efficiency gradients is used as the target multi-functional monitoring data to be processed; when there are target multi-functional monitoring data corresponding to at least three different solar power generation efficiency gradients, the target multi-functional monitoring data of the at least three different solar power generation efficiency gradients is fused to obtain the target multi-functional monitoring data to be processed.

[0009] In this application, each of the energy conversion level prediction models includes a plurality of partial prediction models, and the plurality of partial prediction models correspond to different attributes of the target to be processed; by using the energy conversion level prediction model of the solar power generation efficiency gradient to perform knowledge field extraction processing on the solar power generation monitoring data of the solar energy device to be processed for the solar power generation efficiency gradient, knowledge field information of the target to be processed corresponding to the solar power generation efficiency gradient is obtained, including: for each attribute of the target to be processed, perform the following processing: by using the partial prediction model corresponding to the attribute, perform knowledge field extraction processing on the solar power generation monitoring data of the solar energy device to be processed, and obtain knowledge field information of the target to be processed corresponding to the attribute; by using the energy conversion level prediction model of the solar power generation efficiency gradient and combining the knowledge field information of the target to be processed corresponding to the solar power generation efficiency gradient to perform hidden prediction processing, target multi-functional monitoring data of the target to be processed for the solar power generation efficiency gradient is obtained, including: for each attribute of the target to be processed, perform the following processing: by using the partial prediction model corresponding to the attribute and combining the knowledge field information of the target to be processed corresponding to the attribute to perform hidden prediction processing, target multi-functional monitoring data of the target to be processed corresponding to the attribute is obtained; perform fusion processing on the target multi-functional monitoring data of the target to be processed corresponding to multiple attributes to obtain target multi-functional monitoring data of the target to be processed corresponding to the solar power generation efficiency gradient.

[0010] In this application, the combining the knowledge field information of the target to be processed corresponding to the attribute to perform hidden prediction processing to obtain target multi-functional monitoring data of the target to be processed corresponding to the attribute includes: performing knowledge field deletion processing on the knowledge field information of the target to be processed corresponding to the attribute to obtain the deleted knowledge field information; performing weighting processing on the deleted knowledge field information to obtain target multi-functional monitoring data of the target to be processed corresponding to the attribute.

[0011] In this application, the performing weighting processing on the deleted knowledge field information to obtain target multi-functional monitoring data of the target to be processed corresponding to the attribute includes: mapping the deleted knowledge field information to the target multi-functional monitoring data to obtain an abnormal probability distribution of the monitoring area; using the target multi-functional monitoring data corresponding to the maximum probability in the abnormal probability distribution of the monitoring area as the target multi-functional monitoring data of the target to be processed corresponding to the attribute.

[0012] In this application, before performing the hidden knowledge field extraction process on the power generation monitoring data of the solar energy device to be processed, it further includes: classifying the power generation monitoring data of the solar energy device to be processed to obtain multiple solar energy device abnormal working area data of the power generation monitoring data of the solar energy device to be processed; performing the hidden knowledge field extraction process on the power generation monitoring data of the solar energy device to be processed to obtain the hidden knowledge field information of the target to be processed in the power generation monitoring data of the solar energy device to be processed, including: performing the hidden knowledge field extraction process on each of the solar energy device abnormal working area data to obtain the hidden knowledge field information of the target to be processed in the solar energy device abnormal working area data; performing the prediction process by combining the hidden knowledge field information of the target to be processed to obtain the target multi-functional monitoring data to be processed, including: performing the prediction process through a prediction model and combining the hidden knowledge field information of the target to be processed in each of the solar energy device abnormal working area data to obtain the target multi-functional monitoring data of the target to be processed corresponding to each of the solar energy device abnormal working area data; splicing the target multi-functional monitoring data of the target to be processed corresponding to multiple such ranges to obtain the target multi-functional monitoring data to be processed.

[0013] In this application, splicing the target multi-functional monitoring data of the target to be processed corresponding to multiple such ranges to obtain the target multi-functional monitoring data to be processed includes: performing a fusion process on the target multi-functional monitoring data of the target to be processed corresponding to multiple such ranges to obtain the fused target multi-functional monitoring data; performing an elimination process on the fused target multi-functional monitoring data to obtain the target multi-functional monitoring data to be processed.

[0014] In this application, splicing the target multi-functional monitoring data of the target to be processed corresponding to multiple such ranges to obtain the target multi-functional monitoring data to be processed includes: based on the device aging description data of the target to be processed represented by the target multi-functional monitoring data, determining the target multi-functional monitoring data with the largest device aging description data from the target multi-functional monitoring data of the target to be processed corresponding to multiple solar energy device abnormal working area data, and using the target multi-functional monitoring data representing the largest device aging description data as the target multi-functional monitoring data to be processed.

[0015] In this application, monitoring and processing the fusion result of the target to be processed to obtain a solar power generation monitoring result includes: projecting the fusion result of the target to be processed onto the abnormal probability distribution of the monitoring area of the solar power generation monitoring result; using the solar power generation efficiency gradient corresponding to the maximum probability in the abnormal probability distribution of the monitoring area as the solar power generation monitoring result.

[0016] In this application, before extracting the hidden knowledge fields from the solar power generation monitoring data of the solar energy device to be processed, it further includes: in response to a monitoring instruction for the solar power generation monitoring data of the solar energy device to be processed, obtaining the original solar power generation monitoring data; performing at least one of the following processes on the original solar power generation monitoring data: normalizing each data set in the original solar power generation monitoring data and using the normalized original solar power generation monitoring data as the solar power generation monitoring data of the solar energy device to be processed; deleting the interference information in the original solar power generation monitoring data and using the original solar power generation monitoring data after deletion as the solar power generation monitoring data of the solar energy device to be processed; analyzing the target to be processed in the original solar power generation monitoring data and using the original solar power generation monitoring data after analysis as the solar power generation monitoring data of the solar energy device to be processed.

[0017] In a second aspect, a multifunctional monitoring system for a solar power generation system is provided, including a processor and a memory that communicate with each other. The processor is configured to read and execute a computer program from the memory to implement the above method.

[0018] The multifunctional monitoring method and system for a solar power generation system provided by an embodiment of this application, by fusing the multifunctional monitoring data of the target to be processed with the important knowledge field information of the target to be processed, to synthesize the hidden knowledge field information and the important knowledge field information of the target to be processed, extract more knowledge field information of the target to be processed, and can accurately determine the multifunctional monitoring data, thereby improving the accuracy and reliability of the monitoring of the solar power generation system; in addition, by the multifunctional monitoring data of the target to be processed, monitoring the target to be processed to obtain a solar power generation monitoring result to ensure the accuracy of the monitoring, can ensure the normal operation of the equipment, and thus ensure the stability of the power generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required for the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0020] Figure 1 It is a flowchart of a multi-functional monitoring method for a solar power generation system provided by an embodiment of the present application. Specific embodiments

[0021] To better understand the above technical solutions, the following will make a detailed description of the technical solutions of the present application through the drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific knowledge fields in the embodiments are a detailed description of the technical solutions of the present application, rather than a limitation of the technical solutions of the present application. Without conflict, the embodiments of the present application and the technical knowledge fields in the embodiments can be combined with each other.

[0022] Please refer to Figure 1 , which shows a multi-functional monitoring method for a solar power generation system. The method may include the technical solutions described in the following steps 101-105.

[0023] In step 101, through an energy conversion level prediction model of the solar power generation efficiency gradient, knowledge field extraction processing of the solar power generation monitoring data of the solar energy device to be processed is performed for the solar power generation efficiency gradient, and the knowledge field information of the target corresponding solar power generation efficiency gradient to be processed is obtained as the hidden knowledge field information of the target to be processed.

[0024] Among them, the solar power generation efficiency gradient can be classified into 1-10 levels, and the risk level is higher as the value increases.

[0025] Exemplarily, the solar power generation monitoring data of the solar energy device can be understood as the stability of the rock embedded in the ground (whether it can be fixed) and the stability of the internal structure of the rock (whether the rock will crack or break), etc. For example: in the Three Gorges project, according to the relationship between the fissure and the attitude of the free face, the stereographic projection map is applied, and the stability of the dangerous rock monomer and its secondary dangerous rock mass in the exploration area is macroscopically analyzed in combination with the deformation and failure knowledge field.

[0026] For example, the energy conversion level prediction model is a type of artificial intelligence model that can analyze the probability of rock collapse. Among them, artificial intelligence is an interdisciplinary and emerging discipline that is based on computer science and integrates multiple disciplines such as computer science, psychology, and philosophy. It is a new technical science that studies and develops theories, methods, technologies, and application systems for simulating, extending, and expanding human intelligence, attempting to understand the essence of intelligence and produce an intelligent machine that can respond in a way similar to human intelligence. The research in this field includes robots, speech recognition, image recognition, natural language processing, and expert systems, etc.

[0027] Among them, the knowledge field can be understood as a feature.

[0028] Among them, the hidden knowledge field information can include features that cannot be directly monitored and need to be obtained through analysis.

[0029] As an example of obtaining the power generation monitoring data of solar energy devices that need to be processed, the collected power generation monitoring data of solar energy devices that need to be processed is sent to the terminal, and the terminal forwards the power generation monitoring data of solar energy devices that need to be processed to the server, so that the server extracts the hidden knowledge field of the target that needs to be processed from the power generation monitoring data of solar energy devices that need to be processed, and obtains the hidden knowledge field information of the target that needs to be processed in the power generation monitoring data of solar energy devices that need to be processed. Subsequently, subsequent analysis and processing can be carried out according to the hidden knowledge field information of the target that needs to be processed.

[0030] For some alternative embodiments, before performing the hidden knowledge field extraction process on the solar device power generation monitoring data to be processed, it further includes: in response to a monitoring instruction for the solar device power generation monitoring data to be processed, obtaining the original solar device power generation monitoring data; performing at least one of the following processes on the original solar device power generation monitoring data: normalizing each data set in the original solar device power generation monitoring data, and using the normalized original solar device power generation monitoring data as the solar device power generation monitoring data to be processed; deleting interference information in the original solar device power generation monitoring data, and using the original solar device power generation monitoring data after deletion as the solar device power generation monitoring data to be processed; analyzing the target to be processed in the original solar device power generation monitoring data, and using the original solar device power generation monitoring data after analysis as the solar device power generation monitoring data to be processed. Among them, the normalization process can be understood as nondimensionalization, which means removing some or all of the units of an equation involving physical quantities through a suitable variable substitution for the purpose of simplifying experiments or calculations.

[0031] For example, the original solar device power generation monitoring data is collected, and the terminal forwards the original solar device power generation monitoring data to the server. After receiving the original solar device power generation monitoring data, the server preprocesses the original solar device power generation monitoring data. The preprocessed solar device power generation monitoring data is more suitable for subsequent solar device power generation monitoring data analysis. Among them, the preprocessing can be understood as optimizing the original solar device power generation monitoring data through hidden knowledge fields. For example, normalizing each data set in the original solar device power generation monitoring data, and using the normalized original solar device power generation monitoring data as the solar device power generation monitoring data to be processed. Normalization makes some knowledge fields of the solar device power generation monitoring data have invariant properties under a given transformation; deleting interference information in the original solar device power generation monitoring data, and using the original solar device power generation monitoring data after deletion as the solar device power generation monitoring data to be processed. Deletion can eliminate random interference information in the solar device power generation monitoring data; analyzing the target to be processed in the original solar device power generation monitoring data, and using the original solar device power generation monitoring data after analysis as the solar device power generation monitoring data to be processed. Analysis can specifically analyze the information in the solar device power generation monitoring data to improve the inaccuracy problem of the solar device power generation monitoring data.

[0032] In step 102, predictive processing is performed in combination with the hidden knowledge field information of the target to be processed, and the target multi-functional monitoring data to be processed is obtained.

[0033] For example, after the server obtains the hidden knowledge field information of the target to be processed, predictive processing can be performed based on the hidden knowledge field information of the target to be processed to obtain the target multi-functional monitoring data to be processed, so as to perform subsequent monitoring operations on the solar power generation monitoring data according to the target multi-functional monitoring data to be processed.

[0034] A multi-functional monitoring method for a solar power generation system provided by an embodiment of the present invention, step 101 includes step 1011A: In step 1011A, for any one of the multiple solar power generation efficiency gradients, the following processing is performed: Through the energy conversion level prediction model of the solar power generation efficiency gradient, knowledge field extraction processing of the solar power generation monitoring data to be processed is performed to obtain the knowledge field information of the target corresponding to the solar power generation efficiency gradient to be processed, so as to be used as the hidden knowledge field information of the target to be processed. Further, step 102 includes step 1021A-step 1022A: In step 1021A, through the energy conversion level prediction model of the solar power generation efficiency gradient, and in combination with the knowledge field information of the target corresponding to the solar power generation efficiency gradient to be processed, hidden prediction processing is performed to obtain the target multi-functional monitoring data of the target corresponding to the solar power generation efficiency gradient; in step 1022A, the target multi-functional monitoring data of the target corresponding to multiple solar power generation efficiency gradients to be processed is fused to obtain the target multi-functional monitoring data to be processed.

[0035] It should be understood that the prediction models for predicting solar power generation monitoring data include multiple energy conversion level prediction models, and the multiple energy conversion level prediction models respectively correspond to different solar power generation efficiency gradients; when performing prediction processing in combination with the hidden knowledge field information of the target to be processed, the problem of inaccurate prediction is improved, so as to ensure the accuracy of the target multi-functional monitoring data to be processed.

[0036] Among them, the prediction model for predicting the power generation monitoring data of solar energy devices includes multiple energy conversion level prediction models, and the multiple energy conversion level prediction models respectively correspond to different solar power generation efficiency gradients. For any one of the multiple solar power generation efficiency gradients, through the decision-making unit in the energy conversion level prediction model corresponding to this solar power generation efficiency gradient, the knowledge field information for this solar power generation efficiency gradient in the power generation monitoring data of the solar energy device to be processed is extracted, and combined with the knowledge field information for the target corresponding to this solar power generation efficiency gradient to be processed, the target to be processed is hiddenly predicted through the energy conversion level prediction model corresponding to this solar power generation efficiency gradient, so as to obtain the target multi-functional monitoring data for the target corresponding to this solar power generation efficiency gradient to be processed, that is, targeted target multi-functional monitoring data, and the targeted target multi-functional monitoring data are fused to obtain the target multi-functional monitoring data for all solar power generation efficiency gradients, so as to perform accurate target monitoring according to the target multi-functional monitoring data for all solar power generation efficiency gradients subsequently.

[0037] For some alternative embodiments, the hidden prediction process is combined with the knowledge field information for the target corresponding to the solar power generation efficiency gradient to be processed to obtain the target multi-functional monitoring data for the target corresponding to the solar power generation efficiency gradient to be processed, including: performing knowledge field deletion processing on the knowledge field information for the target corresponding to the solar power generation efficiency gradient to be processed to obtain the deleted knowledge field information; performing weighted processing on the deleted knowledge field information to obtain the target multi-functional monitoring data for the target corresponding to the solar power generation efficiency gradient to be processed. Among them, the weighted processing may include a weight calculation method.

[0038] It should be understood that when the hidden prediction process is combined with the knowledge field information for the target corresponding to the solar power generation efficiency gradient to be processed, the problem of interference is improved, so that the target multi-functional monitoring data for the target corresponding to the solar power generation efficiency gradient to be processed can be accurately obtained.

[0039] Furthermore, through the pooling layer in the energy conversion level prediction model corresponding to this solar power generation efficiency gradient, knowledge field deletion is performed on the knowledge field information for the target corresponding to this solar power generation efficiency gradient to be processed to obtain the deleted knowledge field information, so as to delete unimportant knowledge field information, and through the prediction unit in the energy conversion level prediction model corresponding to this solar power generation efficiency gradient, weighted processing is performed on the deleted knowledge field information, so as to obtain the target multi-functional monitoring data for the target corresponding to this solar power generation efficiency gradient to be processed.

[0040] For some alternative embodiments, the target multi-functional monitoring data corresponding to multiple solar power generation efficiency gradients that need to be processed are fused to obtain the target multi-functional monitoring data that needs to be processed, including: when there is target multi-functional monitoring data corresponding to two different solar power generation efficiency gradients, the distinction between the target multi-functional monitoring data of the two different solar power generation efficiency gradients is used as the target multi-functional monitoring data that needs to be processed; when there is target multi-functional monitoring data corresponding to at least three different solar power generation efficiency gradients, the target multi-functional monitoring data of the at least three different solar power generation efficiency gradients are fused to obtain the target multi-functional monitoring data that needs to be processed.

[0041] It should be understood that when fusing the target multi-functional monitoring data corresponding to multiple solar power generation efficiency gradients that need to be processed, the problem of fusion error is improved, so that the target multi-functional monitoring data that needs to be processed can be accurately obtained.

[0042] Further, when there are two different solar power generation efficiency gradients, target multi-functional monitoring data corresponding to the two different solar power generation efficiency gradients will be generated, and the distinction between the target multi-functional monitoring data of the two different solar power generation efficiency gradients is used as the target multi-functional monitoring data that needs to be processed, so as to extract and learn knowledge fields through contrastive learning, and better extract intra-domain commonalities and inter-domain distinctiveness; when there are at least three different solar power generation efficiency gradients, target multi-functional monitoring data corresponding to the at least three different solar power generation efficiency gradients will be generated, and the target multi-functional monitoring data of the at least three different solar power generation efficiency gradients are fused, and the fusion result is used as the target multi-functional monitoring data that needs to be processed, so that the target multi-functional monitoring data that needs to be processed includes the target multi-functional monitoring data for all solar power generation efficiency gradients, so as to accurately monitor the solar device power generation monitoring data according to the target multi-functional monitoring data that needs to be processed subsequently.

[0043] For some alternative embodiments, the solar power generation efficiency gradient is further subdivided, and through the energy conversion level prediction model of the solar power generation efficiency gradient, knowledge field extraction processing of the solar device power generation monitoring data that needs to be processed is performed to obtain the knowledge field information corresponding to the solar power generation efficiency gradient of the target that needs to be processed, including: for each attribute of the target that needs to be processed, the following processing is performed: through the partial prediction model corresponding to the attribute, knowledge field extraction processing of the solar device power generation monitoring data that needs to be processed is performed to obtain the knowledge field information corresponding to the attribute of the target that needs to be processed; Furthermore, through an energy conversion level prediction model based on the solar power generation efficiency gradient, and combined with the knowledge field information of the target corresponding to the solar power generation efficiency gradient to be processed for hidden prediction processing, the target multi-functional monitoring data of the target to be processed at the solar power generation efficiency gradient is obtained, including: for each attribute of the target to be processed, the following processing is performed: through the partial prediction model corresponding to the attribute, and combined with the knowledge field information of the target corresponding to the attribute to be processed for hidden prediction processing, the target multi-functional monitoring data of the target corresponding to the attribute to be processed is obtained; the target multi-functional monitoring data corresponding to multiple attributes of the target to be processed is fused to obtain the target multi-functional monitoring data of the target to be processed at the solar power generation efficiency gradient.

[0044] Exemplarily, the attribute can be understood as the rock state attribute, such as: the internal structure and composition information of the rock, etc.

[0045] It should be understood that each energy conversion level prediction model includes multiple partial prediction models, and the multiple partial prediction models correspond to different attributes of the target to be processed; when extracting the knowledge field information of the solar power generation efficiency gradient from the solar power generation monitoring data of the solar device to be processed through the energy conversion level prediction model of the solar power generation efficiency gradient, the problem of inaccurate extraction is improved, so that the knowledge field information of the target corresponding to the solar power generation efficiency gradient to be processed can be accurately obtained.

[0046] Furthermore, each energy conversion level prediction model includes multiple partial prediction models, and the multiple partial prediction models correspond to different attributes of the target to be processed; for any one of the multiple attributes at a certain solar power generation efficiency gradient, through the decision-making unit in the partial prediction model corresponding to the attribute, the knowledge field information of the attribute in the solar power generation monitoring data of the solar device to be processed is extracted, and combined with the knowledge field information of the target corresponding to the attribute to be processed, the target to be processed is hidden-predicted through the energy conversion level prediction model corresponding to the attribute to obtain the target multi-functional monitoring data of the target corresponding to the attribute to be processed, that is, targeted target multi-functional monitoring data is obtained, and the targeted target multi-functional monitoring data is fused to obtain the target multi-functional monitoring data for all attributes, and the target multi-functional monitoring data corresponding to multiple attributes of the target to be processed is fused to obtain the target multi-functional monitoring data of the target to be processed at the solar power generation efficiency gradient, so as to perform accurate target monitoring according to the target multi-functional monitoring data for all attributes in the subsequent stage.

[0047] For some alternative embodiments, a hidden prediction process is performed in combination with the knowledge field information of the target corresponding attribute to be processed, and the target multi-functional monitoring data of the target corresponding attribute to be processed is obtained, including: performing a knowledge field deletion process on the knowledge field information of the target corresponding attribute to be processed to obtain the deleted knowledge field information; performing a weighting process on the deleted knowledge field information to obtain the target multi-functional monitoring data of the target corresponding attribute to be processed.

[0048] It should be understood that when performing the hidden prediction process in combination with the knowledge field information of the target corresponding attribute to be processed, the problem of the influence of interference information is improved, so that the target multi-functional monitoring data of the target corresponding attribute to be processed can be accurately obtained.

[0049] Further, through the pooling layer in the partial prediction model corresponding to this attribute, the knowledge field information of the target corresponding attribute to be processed is deleted to obtain the deleted knowledge field information, so as to delete unimportant knowledge field information, and through the prediction unit in the partial prediction model corresponding to this attribute, the deleted knowledge field information is weighted, thereby obtaining the target multi-functional monitoring data of the target corresponding attribute to be processed.

[0050] For some alternative embodiments, performing a weighting process on the deleted knowledge field information to obtain the target multi-functional monitoring data of the target corresponding attribute to be processed, including: mapping the deleted knowledge field information into the target multi-functional monitoring data to obtain the abnormal probability distribution of the monitoring area; using the target multi-functional monitoring data corresponding to the maximum probability in the abnormal probability distribution of the monitoring area as the target multi-functional monitoring data of the target corresponding attribute to be processed.

[0051] It should be understood that when performing the weighting process on the deleted knowledge field information, the problem of inaccurate abnormal probability distribution of the monitoring area is improved, so that the target multi-functional monitoring data of the target corresponding to the attribute to be processed can be accurately obtained.

[0052] Further, the deleted knowledge field information is mapped into the target multi-functional monitoring data to obtain the abnormal probability distribution of the monitoring area, and there are probabilities of multiple different target multi-functional monitoring data corresponding to this attribute in the abnormal probability distribution of the monitoring area.

[0053] A multi-functional monitoring method for a solar power generation system provided by an embodiment of the present invention. In order to accurately obtain target multi-functional monitoring data to be processed, it further includes step 106: In step 106, the power generation monitoring data of solar devices to be processed is classified to obtain multiple solar device abnormal working area data of the power generation monitoring data of solar devices to be processed; then step 101 includes step 1011B: In step 1011B, hidden knowledge field extraction processing is performed on each solar device abnormal working area data to obtain hidden knowledge field information of the target in the solar device abnormal working area data to be processed; among them, the hidden knowledge field extraction processing can be understood as key content extraction or recognition.

[0054] It should be understood that when splicing the target multi-functional monitoring data corresponding to the target to be processed for multiple said ranges, the problem of repetition is improved, so that the target multi-functional monitoring data to be processed can be accurately obtained.

[0055] Further, step 102 includes step 1021B-step 1022B: In step 1021B, prediction processing is performed through a prediction model and in combination with the hidden knowledge field information of the target to be processed in each solar device abnormal working area data to obtain target multi-functional monitoring data corresponding to the target to be processed for each solar device abnormal working area data; in step 1022B, the target multi-functional monitoring data corresponding to the target to be processed for multiple ranges is spliced to obtain the target multi-functional monitoring data to be processed. Among them, prediction can be understood as prediction.

[0056] It should be understood that when performing prediction processing in combination with the hidden knowledge field information of the target to be processed, the problem of inaccurate prediction processing is improved, so that the target multi-functional monitoring data to be processed can be accurately obtained.

[0057] For example, first classify the power generation monitoring data of solar devices that need to be processed to obtain multiple solar device abnormal working area data of the power generation monitoring data of solar devices that need to be processed. Then, through a prediction model for predicting the power generation monitoring data of solar devices, combined with the hidden knowledge field information of the target to be processed in each solar device abnormal working area data, perform a prediction to obtain the target multi-functional monitoring data of the target to be processed corresponding to each solar device abnormal working area data. Finally, integrate the target multi-functional monitoring data of the target to be processed corresponding to multiple ranges to obtain the target multi-functional monitoring data of the target to be processed. By obtaining the partial target multi-functional monitoring data corresponding to each solar device abnormal working area data, accurately obtain all the target multi-functional monitoring data in the power generation monitoring data of solar devices that need to be processed, and avoid missing some of the target multi-functional monitoring data in the power generation monitoring data of solar devices that need to be processed.

[0058] For some alternative embodiments, splice the target multi-functional monitoring data of the target to be processed corresponding to multiple ranges to obtain the target multi-functional monitoring data of the target to be processed, including: perform a fusion process on the target multi-functional monitoring data of the target to be processed corresponding to multiple ranges to obtain the fused target multi-functional monitoring data; perform an elimination process on the fused target multi-functional monitoring data to obtain the target multi-functional monitoring data of the target to be processed.

[0059] Furthermore, after the server obtains the target multi-functional monitoring data of the target to be processed corresponding to each range, it can first fuse the target multi-functional monitoring data of the target to be processed corresponding to multiple ranges (the multiple ranges can be partial ranges among all ranges or all ranges). Since there may be duplicates in the target multi-functional monitoring data of multiple ranges, eliminate the fused target multi-functional monitoring data to remove the duplicate target multi-functional monitoring data and obtain the non-duplicate target multi-functional monitoring data of the target to be processed.

[0060] For some alternative embodiments, splice the target multi-functional monitoring data of the target to be processed corresponding to multiple ranges to obtain the target multi-functional monitoring data of the target to be processed, including: according to the device aging description data of the matter to be processed represented by the target multi-functional monitoring data, determine the target multi-functional monitoring data with the largest device aging description data from the target multi-functional monitoring data of the target to be processed corresponding to multiple solar device abnormal working area data, and use the target multi-functional monitoring data representing the largest device aging description data as the target multi-functional monitoring data of the target to be processed.

[0061] It should be understood that when splicing and processing the target multi-functional monitoring data corresponding to multiple ranges that need to be processed, the problem of the unmeasurability of the device aging description data in the multi-functional monitoring data is improved, so that the target multi-functional monitoring data that needs to be processed can be accurately obtained.

[0062] In step 103, important knowledge field extraction processing is performed on the solar device power generation monitoring data that needs to be processed to obtain the important knowledge field information of the target that needs to be processed.

[0063] Among them, important knowledge field extraction is performed on the solar device power generation monitoring data that needs to be processed to obtain comprehensive knowledge field information of the target that needs to be processed.

[0064] In step 104, the target multi-functional monitoring data that needs to be processed in the solar device power generation monitoring data that needs to be processed is fused with the important knowledge field information of the target that needs to be processed to obtain the fusion result of the target that needs to be processed.

[0065] For example, after the server obtains the target multi-functional monitoring data that needs to be processed and the important knowledge field information of the target that needs to be processed, the target multi-functional monitoring data that needs to be processed and the important knowledge field information of the target that needs to be processed are fused to obtain the fusion result of the target that needs to be processed, so as to perform solar device power generation monitoring data monitoring processing based on the fusion result of the target that needs to be processed subsequently.

[0066] In step 105, monitoring processing is performed in combination with the fusion result of the target that needs to be processed to obtain the solar power generation monitoring result.

[0067] For example, after the server obtains the fusion result of the target that needs to be processed, based on the fusion result of the target that needs to be processed, the target that needs to be processed is monitored to obtain the solar power generation monitoring result.

[0068] For some replaceable embodiments, monitoring processing is performed based on the fusion result of the target that needs to be processed to obtain the solar power generation monitoring result, including: projecting the fusion result of the target that needs to be processed to the abnormal probability distribution of the monitoring area of the solar power generation monitoring result; using the solar power generation efficiency gradient corresponding to the maximum probability in the abnormal probability distribution of the monitoring area as the solar power generation monitoring result.

[0069] It should be understood that when performing monitoring processing in combination with the fusion result of the target that needs to be processed, the problem of unreliable collapse probability is improved, so that the solar power generation monitoring result can be accurately obtained.

[0070] For some alternative embodiments, in order to analyze the solar power generation efficiency gradient of an object through a solar power generation monitoring data processing model, it is necessary to configure the solar power generation monitoring data processing model. The configuration process includes: performing prediction processing on a solar power generation monitoring data example through the solar power generation monitoring data processing model to obtain target multi-functional monitoring data that needs to be processed; fusing the target multi-functional monitoring data that needs to be processed in the solar power generation monitoring data example with the important knowledge field information of the target that needs to be processed to obtain a fusion result of the target that needs to be processed; performing monitoring processing based on the fusion result of the target that needs to be processed to obtain a solar power generation monitoring result; constructing an evaluation index algorithm for the solar power generation monitoring data processing model according to the solar power generation monitoring result and the solar power generation efficiency gradient identifier; updating the coefficients of the solar power generation monitoring data processing model until the evaluation index algorithm converges, and taking the updated coefficients of the solar power generation monitoring data processing model when the evaluation index algorithm converges as the coefficients of the configured solar power generation monitoring data processing model.

[0071] For example, through the solar power generation monitoring data processing model, perform hidden knowledge field extraction processing on the solar power generation monitoring data example to obtain the hidden knowledge field information of the target that needs to be processed in the solar power generation monitoring data example. Based on the hidden knowledge field information of the target that needs to be processed, perform prediction processing to obtain the target multi-functional monitoring data that needs to be processed. Perform important knowledge field extraction processing on the solar power generation monitoring data example to obtain the important knowledge field information of the target that needs to be processed. Fuse the target multi-functional monitoring data that needs to be processed in the solar power generation monitoring data example with the important knowledge field information of the target that needs to be processed to obtain a fusion result of the target that needs to be processed. Perform monitoring processing based on the fusion result of the target that needs to be processed to obtain a solar power generation monitoring result. According to the solar power generation monitoring result and the solar power generation efficiency gradient identifier, after determining the value of the evaluation index algorithm of the solar power generation monitoring data processing model, it can be judged whether the value of the evaluation index algorithm of the solar power generation monitoring data processing model exceeds the specified target value. When the value of the evaluation index algorithm of the solar power generation monitoring data processing model exceeds the specified target value, determine the error information of the solar power generation monitoring data processing model based on the evaluation index algorithm of the solar power generation monitoring data processing model, input the error information into the solar power generation monitoring data processing model for optimization, and update each model coefficient during the optimization process.

[0072] On this basis, a multi-functional monitoring device for a solar power generation system is provided. The device includes: An information determination module, configured to perform knowledge field extraction processing of the solar power generation efficiency gradient on the power generation monitoring data of the solar energy device to be processed, so as to obtain the knowledge field information corresponding to the target to be processed for the solar power generation efficiency gradient, and use it as the hidden knowledge field information of the target to be processed; An information prediction module, configured to perform prediction processing by combining the hidden knowledge field information of the target to be processed, so as to obtain the multi-functional monitoring data of the target to be processed; A knowledge field extraction module, configured to perform important knowledge field extraction processing on the power generation monitoring data of the solar energy device to be processed, so as to obtain the important knowledge field information of the target to be processed; A result fusion module, configured to perform fusion processing on the multi-functional monitoring data of the target to be processed in the power generation monitoring data of the solar energy device to be processed and the important knowledge field information of the target to be processed, so as to obtain the fusion result of the target to be processed; A result monitoring module, configured to perform monitoring processing by combining the fusion result of the target to be processed, so as to obtain the solar power generation monitoring result.

[0073] On this basis, a multi-functional monitoring system for a solar power generation system is shown, including a processor and a memory that communicate with each other. The processor is configured to read and execute a computer program from the memory to implement the above method.

[0074] On this basis, a computer-readable storage medium is further provided, and the computer program stored thereon implements the above method when running.

[0075] In summary, based on the above solution, by fusing the multi-functional monitoring data of the target to be processed with the important knowledge field information of the target to be processed, and integrating the hidden knowledge field information of the target to be processed with the important knowledge field information of the target to be processed, more knowledge field information of the target to be processed can be extracted, and the multi-functional monitoring data can be accurately determined, thereby improving the accuracy and reliability of the solar power generation system monitoring; in addition, through the multi-functional monitoring data of the target to be processed, the target to be processed is monitored to obtain the solar power generation monitoring result, so as to ensure the accuracy of the monitoring, ensure the normal operation of the equipment, and thus ensure the stability of the power generation.

[0076] It should be understood that the systems and their modules shown above can be implemented in various ways. For example, in some embodiments, the systems and their modules can be implemented by hardware, software, or a combination of software and hardware. Among them, the hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated designed hardware. Those skilled in the art can understand that the above methods and systems can be implemented using computer-executable instructions and / or included in processor control code. For example, such code is provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The systems and their modules of the present application can be implemented not only by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips and transistors, or programmable hardware devices such as field programmable gate arrays and programmable logic devices, but also by software implemented by various types of processors, or by a combination of the above hardware circuits and software (e.g., firmware).

[0077] It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the possible beneficial effects can be any one or several combinations of the above, or any other possible beneficial effects that can be obtained.

Claims

1. A multi-functional monitoring method for a solar power generation system, characterized in that, The method includes: Performing knowledge field extraction processing on the power generation monitoring data of the solar energy device to be processed through the energy conversion level prediction model of the solar power generation efficiency gradient, obtaining the knowledge field information of the target to be processed corresponding to the solar power generation efficiency gradient, and using it as the hidden knowledge field information of the target to be processed; Performing prediction processing in combination with the hidden knowledge field information of the target to be processed to obtain the multi-functional monitoring data of the target to be processed; Performing important knowledge field extraction processing on the power generation monitoring data of the solar energy device to be processed to obtain the important knowledge field information of the target to be processed; Performing fusion processing on the multi-functional monitoring data of the target to be processed in the power generation monitoring data of the solar energy device to be processed and the important knowledge field information of the target to be processed to obtain the fusion result of the target to be processed; Performing monitoring processing in combination with the fusion result of the target to be processed to obtain the solar power generation monitoring result.

2. The method according to claim 1, wherein The prediction model for predicting the power generation monitoring data of the solar energy device includes multiple energy conversion level prediction models, and the multiple energy conversion level prediction models respectively correspond to different solar power generation efficiency gradients; the performing prediction processing in combination with the hidden knowledge field information of the target to be processed to obtain the multi-functional monitoring data of the target to be processed includes: Through the energy conversion level prediction model of the solar power generation efficiency gradient, and performing hidden prediction processing in combination with the knowledge field information of the target to be processed corresponding to the solar power generation efficiency gradient, obtaining the multi-functional monitoring data of the target to be processed corresponding to the solar power generation efficiency gradient; Performing fusion processing on the multi-functional monitoring data of the target to be processed corresponding to multiple solar power generation efficiency gradients to obtain the multi-functional monitoring data of the target to be processed.

3. The method according to claim 2, wherein The performing hidden prediction processing in combination with the knowledge field information of the target to be processed corresponding to the solar power generation efficiency gradient to obtain the multi-functional monitoring data of the target to be processed corresponding to the solar power generation efficiency gradient includes: Performing knowledge field deletion processing on the knowledge field information of the target to be processed corresponding to the solar power generation efficiency gradient to obtain the deleted knowledge field information; Performing weighting processing on the deleted knowledge field information to obtain the multi-functional monitoring data of the target to be processed corresponding to the solar power generation efficiency gradient.

4. The method according to claim 2, wherein The performing fusion processing on the multi-functional monitoring data of the target to be processed corresponding to multiple solar power generation efficiency gradients to obtain the multi-functional monitoring data of the target to be processed includes: When there are multi-functional monitoring data corresponding to two different solar power generation efficiency gradients, taking the distinction between the multi-functional monitoring data of the two different solar power generation efficiency gradients as the multi-functional monitoring data of the target to be processed; When there is target multi-functional monitoring data corresponding to at least three different solar power generation efficiency gradients, perform fusion processing on the target multi-functional monitoring data of the at least three different solar power generation efficiency gradients to obtain the target multi-functional monitoring data that needs to be processed.

5. The method according to claim 2, wherein Each of the energy conversion level prediction models includes a plurality of partial prediction models, and the plurality of partial prediction models correspond to different attributes of the target that needs to be processed; through the energy conversion level prediction model of the solar power generation efficiency gradient, perform knowledge field extraction processing on the solar power generation monitoring data of the solar energy device that needs to be processed to obtain the knowledge field information of the target that needs to be processed corresponding to the solar power generation efficiency gradient, including: For each attribute of the target that needs to be processed, perform the following processing: through the partial prediction model corresponding to the attribute, perform knowledge field extraction processing on the solar power generation monitoring data of the solar energy device that needs to be processed to obtain the knowledge field information of the target that needs to be processed corresponding to the attribute; Through the energy conversion level prediction model of the solar power generation efficiency gradient, and combining the knowledge field information of the target that needs to be processed corresponding to the solar power generation efficiency gradient to perform hidden prediction processing to obtain the target multi-functional monitoring data of the target that needs to be processed at the solar power generation efficiency gradient, including: For each attribute of the target that needs to be processed, perform the following processing: through the partial prediction model corresponding to the attribute, and combining the knowledge field information of the target that needs to be processed corresponding to the attribute to perform hidden prediction processing to obtain the target multi-functional monitoring data of the target that needs to be processed corresponding to the attribute; Perform fusion processing on the target multi-functional monitoring data of the target that needs to be processed corresponding to multiple attributes to obtain the target multi-functional monitoring data of the target that needs to be processed corresponding to the solar power generation efficiency gradient.

6. The method according to claim 5, characterized in that, The combining the knowledge field information of the target that needs to be processed corresponding to the attribute to perform hidden prediction processing to obtain the target multi-functional monitoring data of the target that needs to be processed corresponding to the attribute includes: Perform knowledge field deletion processing on the knowledge field information of the target that needs to be processed corresponding to the attribute to obtain the deleted knowledge field information; Perform weighting processing on the deleted knowledge field information to obtain the target multi-functional monitoring data of the target that needs to be processed corresponding to the attribute; Among them, the performing weighting processing on the deleted knowledge field information to obtain the target multi-functional monitoring data of the target that needs to be processed corresponding to the attribute includes: Map the deleted knowledge field information into the target multi-functional monitoring data to obtain the abnormal probability distribution of the monitoring area; Use the target multi-functional monitoring data corresponding to the maximum probability in the abnormal probability distribution of the monitoring area as the target multi-functional monitoring data of the target that needs to be processed corresponding to the attribute.

7. The method according to claim 1, wherein Before performing the hidden knowledge field extraction process on the power generation monitoring data of the solar energy device to be processed, it further includes: classifying the power generation monitoring data of the solar energy device to be processed to obtain multiple solar energy device abnormal working area data of the power generation monitoring data of the solar energy device to be processed; Performing the hidden knowledge field extraction process on the power generation monitoring data of the solar energy device to be processed to obtain the hidden knowledge field information of the target to be processed in the power generation monitoring data of the solar energy device to be processed, including: performing the hidden knowledge field extraction process on each of the solar energy device abnormal working area data to obtain the hidden knowledge field information of the target to be processed in the solar energy device abnormal working area data; Combining the hidden knowledge field information of the target to be processed to perform a prediction process to obtain the target multi-functional monitoring data of the target to be processed, including: through a prediction model, and combining the hidden knowledge field information of the target to be processed in each of the solar energy device abnormal working area data to perform a prediction process to obtain the target multi-functional monitoring data of the target to be processed corresponding to each of the solar energy device abnormal working area data; splicing the target multi-functional monitoring data of the target to be processed corresponding to multiple ranges to obtain the target multi-functional monitoring data of the target to be processed; Among them, the splicing process of the target multi-functional monitoring data of the target to be processed corresponding to multiple ranges to obtain the target multi-functional monitoring data of the target to be processed includes: Performing a fusion process on the target multi-functional monitoring data of the target to be processed corresponding to multiple ranges to obtain the fused target multi-functional monitoring data; Performing an elimination process on the fused target multi-functional monitoring data to obtain the target multi-functional monitoring data of the target to be processed; Among them, the splicing process of the target multi-functional monitoring data of the target to be processed corresponding to multiple ranges to obtain the target multi-functional monitoring data of the target to be processed includes: according to the device aging description data of the target to be processed represented by the target multi-functional monitoring data, determining the target multi-functional monitoring data with the largest device aging description data from the target multi-functional monitoring data of the target to be processed corresponding to multiple solar energy device abnormal working area data, and using the target multi-functional monitoring data representing the largest device aging description data as the target multi-functional monitoring data of the target to be processed.

8. The method according to claim 1, characterized in that, Combining the fusion result of the target to be processed to perform a monitoring process to obtain the solar power generation monitoring result, including: Projecting the fusion result of the target to be processed onto the monitoring area abnormal probability distribution of the solar power generation monitoring result; Using the solar power generation efficiency gradient corresponding to the maximum probability in the monitoring area abnormal probability distribution as the solar power generation monitoring result.

9. The method according to claim 1, characterized in that Before performing the processing of extracting hidden knowledge fields from the power generation monitoring data of the solar energy device to be processed, the following steps are further included: In response to a monitoring instruction for the power generation monitoring data of the solar energy device to be processed, obtain the original power generation monitoring data of the solar energy device; perform at least one of the following processes on the original power generation monitoring data of the solar energy device: perform normalization processing on each data set in the original power generation monitoring data of the solar energy device, and use the normalized original power generation monitoring data of the solar energy device as the power generation monitoring data of the solar energy device to be processed; Delete the interference information in the original power generation monitoring data of the solar energy device, and use the original power generation monitoring data after deletion as the power generation monitoring data of the solar energy device to be processed; Analyze the target to be processed in the original power generation monitoring data of the solar energy device, and use the original power generation monitoring data after analysis as the power generation monitoring data of the solar energy device to be processed.

10. A multi-functional monitoring system for a solar power generation system, characterized in that, It includes a processor and a memory that communicate with each other. The processor is configured to read and execute a computer program from the memory to implement the method according to any one of claims 1-9.

Images