CN113888844A

CN113888844A - Shielding early warning method and device for photovoltaic power generation equipment and electronic equipment

Info

Publication number: CN113888844A
Application number: CN202111088274.7A
Authority: CN
Inventors: 杜雪峰; 聂宏涛; 王门麟
Original assignee: Jiangsu Jinfeng Software Technology Co ltd; Beijing Goldwind Smart Energy Service Co Ltd
Current assignee: Jiangsu Jinfeng Software Technology Co ltd; Beijing Goldwind Smart Energy Service Co Ltd
Priority date: 2021-09-16
Filing date: 2021-09-16
Publication date: 2022-01-04
Anticipated expiration: 2041-09-16
Also published as: CN113888844B

Abstract

The application discloses a shielding early warning method and device for photovoltaic power generation equipment and electronic equipment. According to the embodiment of the application, the relevant data of the target electric field are obtained, wherein the relevant data comprise environmental data of the target electric field, operation data of target power generation equipment in the target electric field and operation data of reference power generation equipment in the target electric field; inputting the related data into an occlusion identification model to obtain an occlusion identification result, wherein the occlusion identification result is used for representing whether occlusion exists in the target power generation equipment, the occlusion identification model comprises a first sub-model and a second sub-model, the first sub-model is used for identifying short-term occlusion, and the second sub-model is used for identifying long-term occlusion; early warning information is output according to the occlusion recognition result, the early warning information is used for indicating whether the target power generation equipment has short-term occlusion and long-term occlusion, and the technical problem that whether the photovoltaic power generation equipment has occlusion cannot be accurately judged by an occlusion early warning method in the related technology can be solved.

Description

Shielding early warning method and device for photovoltaic power generation equipment and electronic equipment

Technical Field

The application belongs to the technical field of photovoltaic power generation, and particularly relates to a shielding early warning method and device for photovoltaic power generation equipment and electronic equipment.

Background

When the photovoltaic power station is designed, whether buildings and trees are shielded around can be considered, but whether new trees, weeds and buildings exist around cannot be predicted after the photovoltaic power station is built. In addition, random factors such as fallen leaves and bird droppings can be caused. In the prior art, a shielding early warning method for determining whether shielding exists according to the change of the generated energy is provided, but due to numerous factors influencing the generated energy, whether the change of the generated energy is caused by shielding cannot be well judged, so that the shielding early warning method in the prior art cannot accurately judge whether the photovoltaic power generation equipment has shielding.

Disclosure of Invention

The embodiment of the application provides a shielding early warning method and device for photovoltaic power generation equipment and electronic equipment, and can solve the technical problem that whether the photovoltaic power generation equipment is shielded or not cannot be accurately judged by the shielding early warning method in the related technology.

In a first aspect, an embodiment of the present application provides a blocking early warning method for a photovoltaic power generation device, where the method includes:

acquiring relevant data of a target electric field, wherein the relevant data comprises environmental data of the target electric field, operation data of target power generation equipment in the target electric field and operation data of reference power generation equipment in the target electric field;

inputting the related data into an occlusion identification model to obtain an occlusion identification result, wherein the occlusion identification result is used for representing whether occlusion exists in the target power generation equipment, the occlusion identification model comprises a first sub-model and a second sub-model, the first sub-model is used for identifying short-term occlusion, and the second sub-model is used for identifying long-term occlusion;

and outputting early warning information according to the shielding identification result, wherein the early warning information is used for indicating whether the target power generation equipment has short-term shielding and long-term shielding.

Further, inputting the relevant data into an occlusion recognition model to obtain an occlusion recognition result, including:

and inputting the operation data of the target power generation equipment and the operation data of the reference power generation equipment into the first sub-model, and comparing the difference between the target power generation equipment and the reference power generation equipment to obtain a shielding identification result of short-term shielding.

Further, inputting the operation data of the target power generation equipment and the operation data of the reference power generation equipment into the first sub-model, and comparing the difference between the target power generation equipment and the reference power generation equipment to obtain a shielding identification result of short-term shielding, wherein the shielding identification result comprises:

respectively calculating branch average power generation parameters of the target power generation equipment and the reference power generation equipment in different time periods of each day according to the operation data of the target power generation equipment and the reference power generation equipment;

and comparing the difference of the branch average power generation parameters of the target power generation equipment and the reference power generation equipment in each time period through the first submodel to judge whether the target power generation equipment has short-term shielding in the corresponding time period.

Further, comparing the difference of the branch average power generation parameters of the target power generation equipment and the reference power generation equipment in each time period through the first submodel to judge whether the target power generation equipment has short-term occlusion in the corresponding time period, including:

and under the condition that the first submodel judges that the average branch power generation parameters of the target power generation equipment and the reference power generation equipment are different in the same time period exceeding m days, determining that the target power generation equipment has short-term shielding, wherein m is a positive integer greater than or equal to 1.

Further, the operation data of the target power generation device includes actual power generation data;

inputting the related data into the occlusion recognition model to obtain an occlusion recognition result, comprising:

inputting the environmental data into a second submodel to obtain predicted power generation data of the target power generation equipment;

and comparing the actual power generation data of the target power generation equipment with the predicted power generation data to obtain a shielding identification result of the long-term shielding.

Further, the target electric field comprises a plurality of target power generation devices;

comparing the actual power generation data and the predicted power generation data of the target power generation equipment to obtain a shielding identification result of long-term shielding, comprising:

calculating error parameters of actual power generation data and predicted power generation data of each target power generation device;

and under the condition that the number of the target power generation devices with the error parameters exceeding the preset error threshold does not exceed the preset number, determining that the target power generation devices with the error parameters exceeding the preset error threshold have long-term shielding.

Further, after acquiring the data related to the electric field of the target, the method further includes:

respectively removing invalid data corresponding to invalid conditions corresponding to the data types from each data of the related data of the target electric field to obtain valid data;

inputting the related data into the occlusion recognition model to obtain an occlusion recognition result, comprising: and inputting the effective data into the occlusion recognition model to obtain an occlusion recognition result.

Further, the operational data includes branch power generation parameters for each branch per day;

the method for removing invalid data corresponding to an invalid condition corresponding to a data type from each data related to a target electric field includes:

and removing the operation data of the branch circuit of which the branch circuit power generation parameter is out of the first parameter range from the operation data of the target power generation equipment and the operation data of the reference power generation equipment respectively.

Further, before removing the operation data of the branch whose branch power generation parameter is out of the first parameter range from the operation data of the target power generation device, the method further includes:

determining the date of the limited operation of the target power generation equipment according to the daily power limit control instruction to obtain a first date;

determining the date of which the daily irradiance is out of the second parameter range to obtain a second date; wherein the environmental data of the target electric field comprises daily irradiance;

removing the operation data of the target power generation equipment on the first date and the second date;

in the operation data of the reference power generation equipment, before removing the operation data of the branch whose branch power generation parameter is out of the first parameter range, the method further comprises the following steps:

the operating data of the reference power generating equipment on the first date and the second date is removed.

Further, the early warning information includes early warning grade, according to sheltering from the output early warning information of discernment result, includes:

responding to the short-term occlusion existence of the occlusion identification result, and outputting a first-level early warning level and a short-term occlusion time period;

responding to the occlusion identification result that long-term occlusion exists, and outputting the early warning level as a second level;

responding to the occlusion identification result that short-term occlusion and long-term occlusion exist at the same time, and outputting a third-level early warning level and a short-term occlusion period;

wherein, the grades of the first grade, the second grade and the third grade are from low to high.

Further, the first sub-model comprises a back propagation BP neural network and the second sub-model comprises a long-short term memory LSTM neural network.

In a second aspect, an embodiment of the present application provides a blocking early warning device for photovoltaic power generation equipment, the device includes:

the device comprises a first acquisition unit, a second acquisition unit and a control unit, wherein the first acquisition unit is used for acquiring relevant data of a target electric field, and the relevant data comprises environmental data of the target electric field, operation data of target power generation equipment in the target electric field and operation data of reference power generation equipment in the target electric field;

the shielding identification model comprises a first submodel and a second submodel, wherein the first submodel is used for identifying short-term shielding, and the second submodel is used for identifying long-term shielding;

and the output unit is used for outputting early warning information according to the shielding identification result, and the early warning information is used for indicating whether the target power generation equipment has short-term shielding and long-term shielding.

Further, the input unit is further configured to input the operation data of the target power generation device and the operation data of the reference power generation device into the first sub-model, and compare a difference between the target power generation device and the reference power generation device to obtain an occlusion recognition result for short-term occlusion.

Further, the input unit includes:

the first calculating subunit is used for respectively calculating branch average power generation parameters of the target power generation equipment and the reference power generation equipment in different time periods each day according to the operation data of the target power generation equipment and the reference power generation equipment;

and the first comparison subunit is used for comparing the difference of the branch average power generation parameters of the target power generation equipment and the reference power generation equipment in each time period through the first submodel to judge whether the target power generation equipment has short-term shielding in the corresponding time period.

Further, the first comparison subunit is further configured to determine that short-term occlusion exists in the target power generation device when the first submodel determines that the target power generation device and the reference power generation device are in the same time period exceeding m days and the branch average power generation parameter is different, where m is a positive integer greater than or equal to 1.

the input unit includes:

the input sub-unit is used for inputting the environment data into the second sub-model to obtain the predicted power generation data of the target power generation equipment;

and the second comparison subunit is used for comparing the actual power generation data of the target power generation equipment with the predicted power generation data to obtain a shielding identification result of the long-term shielding.

the second ratio subunit comprises:

the second calculating subunit is used for calculating error parameters of the actual power generation data and the predicted power generation data of each target power generation device;

the determining subunit is configured to determine that the target power generation device with the error parameter exceeding the preset error threshold has long-term occlusion when the number of the target power generation devices with the error parameter exceeding the preset error threshold does not exceed the preset number.

Further, the apparatus further comprises:

a deleting unit, configured to remove, after acquiring the relevant data of the target electric field, invalid data that conforms to an invalid condition corresponding to the data type from each data of the relevant data of the target electric field, respectively, to obtain valid data;

the input unit is also used for inputting the effective data into the occlusion recognition model to obtain an occlusion recognition result.

the deleting unit is further used for removing the operation data of the branch circuit with the branch circuit power generation parameter out of the first parameter range from the operation data of the target power generation equipment and the operation data of the reference power generation equipment respectively.

Further, the apparatus further comprises:

the first determining unit is used for determining the date of the target power generation equipment which is limited to operate according to the daily electricity limiting control instruction before the operation data of the branch circuit of which the branch circuit power generation parameter is out of the first parameter range is removed from the operation data of the target power generation equipment to obtain a first date;

the second determining unit is used for determining the date of which the daily irradiance is out of the second parameter range to obtain a second date; wherein the environmental data of the target electric field comprises daily irradiance;

the deleting unit is also used for removing the operation data of the target power generation equipment on the first date and the second date, and removing the operation data of the reference power generation equipment on the first date and the second date before removing the operation data of the branch with the branch power generation parameter out of the first parameter range in the operation data of the reference power generation equipment.

Further, the early warning information includes an early warning level, and the output unit includes:

the first output subunit is used for responding to the short-term occlusion existence of the occlusion identification result, and outputting the early warning level as a first level and a short-term occlusion time period;

the second output subunit is used for responding to the occlusion identification result that long-term occlusion exists and outputting the early warning level as a second level;

the third output subunit is used for responding to the occlusion identification result that short-term occlusion and long-term occlusion exist at the same time, and outputting the early warning level as a third level and a short-term occlusion time period;

In another aspect, an embodiment of the present application provides an electronic device, including: a processor and a memory storing computer program instructions; the processor, when executing the computer program instructions, implements the occlusion warning method for a photovoltaic power generation device as provided in the first aspect and any one of its alternative embodiments.

In another aspect, an embodiment of the present application provides a storage medium, where computer program instructions are stored on the storage medium, and when the computer program instructions are executed by a processor, the occlusion warning method for a photovoltaic power generation device, as provided in the first aspect and any optional implementation manner thereof, is implemented.

According to the shielding early warning method, the shielding early warning device, the electronic device and the storage medium of the photovoltaic power generation equipment, whether long-term shielding and short-term shielding exist can be determined respectively based on different submodels used for identifying the long-term shielding and the short-term shielding respectively through the environmental data of a target electric field, the operation data of the target power generation equipment in the target electric field, the operation data of reference power generation equipment in the target electric field and the like, and then early warning information is output according to a shielding identification result.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments of the present application will be briefly described below, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a schematic flowchart of a blocking early warning method for a photovoltaic power generation device according to an embodiment of the present disclosure;

fig. 2 is a schematic flowchart of a blocking warning method for a photovoltaic power generation apparatus according to another embodiment of the present disclosure;

fig. 3 is a schematic flowchart of a blocking warning method for a photovoltaic power generation apparatus according to another embodiment of the present disclosure;

fig. 4 is a schematic flowchart of a blocking warning method for a photovoltaic power generation apparatus according to another embodiment of the present application;

fig. 5 is a schematic structural diagram of a blocking early warning device of a photovoltaic power generation apparatus according to an embodiment of the present application;

fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

Features and exemplary embodiments of various aspects of the present application will be described in detail below, and in order to make objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are intended to be illustrative only and are not intended to be limiting. It will be apparent to one skilled in the art that the present application may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the present application by illustrating examples thereof.

It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

In order to solve the prior art problems, the embodiment of the application provides a shielding early warning method, a shielding early warning device, shielding early warning equipment and a storage medium for photovoltaic power generation equipment. First, a blocking early warning method for photovoltaic power generation equipment provided by the embodiment of the application is introduced below.

Fig. 1 shows a schematic flow chart of a blocking early warning method for photovoltaic power generation equipment according to an embodiment of the present application. As shown in fig. 1, the method comprises the steps of:

step 101, acquiring relevant data of a target electric field.

The relevant data includes environmental data of the target electric field, operation data of the target power generation equipment in the target electric field, and operation data of the reference power generation equipment in the target electric field.

The target electric field is an electric field to be detected for the presence of a blockage. The target electric field includes a target power generation device and a reference power generation device. The target power generation equipment is photovoltaic power generation equipment for detecting whether the shelter exists or not. The reference power generation equipment is photovoltaic power generation equipment used as power generation quantity reference, can be regarded as benchmark equipment in a target electric field, and can be arranged at a position where shielding is least prone to exist in the target electric field.

The environmental data of the target electric field may include daily irradiance monitored within the target electric field, weather conditions, and the like. Illustratively, daily irradiance may be monitored by an irradiance detector disposed in the target electric field; the weather condition can be monitored in the target electric field, and can also be acquired by acquiring weather forecast data of an area where the target electric field is located, so that the weather condition in the target electric field is acquired by acquiring the weather forecast data, and the cost of equipment for monitoring the weather in the target electric field can be reduced.

The operation data of the target power Generation device may include total voltage and total current generated by the target power Generation device every day, voltage and current generated by each branch circuit every day, proportion of an effective photovoltaic string in the device, whether a power limit instruction of automatic power Generation control (agc) is received, and the like. The AGC system is used for regulating whether power generation equipment in a target electric field generates power according to the power consumption peak-valley condition, so that the voltage instability of a power transmission line is avoided, if the power generation equipment receives a power limiting instruction, the power is limited on the same day, the specific power limiting time period can be determined according to the power limiting instruction, and the details are not repeated. The operation data of the reference power generation device is the same as the operation data of the target power generation device, and is not described herein again.

And 102, inputting the related data into an occlusion recognition model to obtain an occlusion recognition result.

And the occlusion recognition result is used for representing whether the target power generation equipment has occlusion or not. The occlusion recognition model is a pre-trained model and comprises a first sub-model and a second sub-model, wherein the first sub-model is used for recognizing short-term occlusion, and the second sub-model is used for recognizing long-term occlusion.

Short term occlusion means occlusion by short term occlusion during only a part of the day, which occlusion typically lasts several hours per day, e.g. may be the shadow of surrounding buildings, or tree weeds, etc.

Long-term shading refers to shading substantially all the time during a day, for example, covering such as dust, bird droppings, leaves, or other sundries on the surface of the device, which if not clear, will shade the photovoltaic device all the time during a day, affecting the power generation capacity of the photovoltaic device.

The shielding identification model in the embodiment of the application can respectively determine whether long-term shielding and short-term shielding exist through different submodels for identifying long-term shielding and short-term shielding, so that whether shielding exists in the photovoltaic power generation equipment can be accurately identified.

Alternatively, the first sub-model may comprise a back propagation BP neural network and the second sub-model may comprise a long-short term memory LSTM neural network. Therefore, the neural network model can be trained through the sample data, and therefore the model capable of accurately identifying the shielding is obtained.

And 103, outputting early warning information according to the shielding identification result.

The early warning information is used for indicating whether short-term occlusion and long-term occlusion exist in the target power generation equipment.

Optionally, the warning information may include a warning level, and in this case, the warning information may be output according to the following three cases:

(1) responding to the short-term occlusion existence of the occlusion identification result, and outputting a first-level early warning level and a short-term occlusion time period;

(2) responding to the occlusion identification result that long-term occlusion exists, and outputting the early warning level as a second level;

(3) and responding to the occlusion identification result that short-term occlusion and long-term occlusion exist at the same time, and outputting the early warning level as a third level and a short-term occlusion period.

For example, the sent warning information may specifically be:

(1) short-term shielding early warning:

early warning grade: is low in

Early warning content: XXX inverter noon time period exist and be sheltered from the foreign matter, please check that whether equipment periphery has sheltering from thing such as trees, wire pole, building, rail, weeds, stone.

(2) And (3) long-term shielding early warning:

early warning grade: in

Early warning content: XXX inverters have a blinding or dirty condition, please check if the device is covered with dust, bird droppings, leaf coverings.

(3) Long-term and short-term occlusions coexist:

early warning grade: height of

Early warning content: the XXX inverter is dirty and is shielded by foreign matters in the midday period to seriously affect the generating capacity, please check whether the equipment is covered by coverings such as dust, bird droppings, leaves and the like, and whether shelters such as trees, telegraph poles, buildings, fences, weeds, stones and the like exist around the equipment.

As shown in fig. 2, which is an optional implementation manner of the embodiment of the present application, the method provided by the embodiment of the present application may be run on a big data platform, and the big data platform may obtain relevant data of a target electric field and store an occlusion recognition model. Through the big data platform, the model can be operated on the cloud platform. The traditional model is operated on several local servers, and cannot provide services for a large number of early warning devices. When a neural network or a deep neural network is used, the method has no way of fast response or even normal operation. And the cloud platform has larger computing resources, and can provide support for the operation of a large amount of equipment and complex models.

The big data platform can provide a uniform scheduling port for each power generation device, generate and send a scheduling task, and the scheduling task is used for indicating whether each power generation device in a target electric field is shielded or not to be detected. After the scheduling task starts, the big data platform can simultaneously run a short-term shielding identification model (a first sub-model) and a long-term shielding identification model (a second sub-model), and the running data of each power generation device is obtained through the scheduling port and input into the sub-models. After the two sub-models run, the program of the big data platform can fuse the judgment results of the two sub-models to generate a final early warning result, then the corresponding maintenance strategy is returned according to the difference of the early warning result, and an operation and maintenance suggestion list is generated for each device. Alternatively, in one application scenario, the maintenance policy may be sent by the big data platform to the SPHM (device health management operating system) in the field. And the field operation and maintenance personnel can check the shielding condition according to the scheme and feed back whether the early warning result is accurate or not. And according to the fed-back field inspection result, each submodel can be subjected to iterative optimization again, and model parameters in the submodels are trained.

The following describes exemplary embodiments of each submodel separately.

For the first sub-model, the operation data of the target power generation device and the operation data of the reference power generation device may be input to the first sub-model, and the difference between the target power generation device and the reference power generation device is compared to obtain the occlusion recognition result for the short-term occlusion. That is, the method for identifying short-term occlusion mainly determines whether some short-term shadow occlusion exists or not by comparing the target power generation device with the reference power generation device.

Further, the above alternative embodiment may include steps 1021 to 1022 as follows:

and 1021, respectively calculating branch average power generation parameters of the target power generation equipment and the reference power generation equipment in different time periods every day according to the operation data of the target power generation equipment and the reference power generation equipment.

The short term occlusion is mainly aimed at the occlusion of building or tree weeds. This type of occlusion usually lasts for several hours, so the pre-warning frequency of the model need not be particularly frequent. For convenience of maintenance of field personnel, a day can be divided into three periods of morning, noon and afternoon. The relevant data used by the sub-model may include environmental data of the environmental monitor (irradiance), operational data of the target power generation device (current, voltage, power generation), operational data of the reference power generation device. The operation Data may be collected by an SCADA (Supervisory Control And Data Acquisition, i.e., Data collection And monitoring Control system) system.

And 1022, comparing the difference of the branch average power generation parameters of the target power generation equipment and the reference power generation equipment in each time period through the first sub-model to judge whether the target power generation equipment has short-term occlusion in the corresponding time period.

When judging whether the photovoltaic power generation equipment has short-term shielding, all data are not needed, the short-term shielding model firstly removes data of electricity limiting days of a power limiting command so as to remove the influence of the electricity limiting command on the power generation amount, and then the days with the optimal power generation amount can be obtained, wherein the days are taken as representatives.

For example, when acquiring relevant data of one month to identify occlusion, only three days in which electricity is not limited in one month and the amount of electricity generated is within a certain range of intervals may be acquired. When the target power generation equipment is judged to have occlusion in a certain period of time based on the data of the three days, the short-term occlusion problem of the power generation equipment can be shown.

Specifically, the determination conditions for determining that there is one example of the short-term occlusion problem are: and under the condition that the first submodel judges that the average branch power generation parameters of the target power generation equipment and the reference power generation equipment are different in the same time period exceeding m days, determining that the target power generation equipment has short-term shielding, wherein m is a positive integer greater than or equal to 1.

Based on the above embodiments, an example of the first sub-model identifying short-term occlusions is provided, and a flowchart is shown in fig. 3.

Referring to fig. 3, first, the obtained related data may specifically include: acquiring monthly environmental monitor data, comprising: irradiance data; the method for acquiring the operation data of the photovoltaic string of the target power generation equipment for one month comprises the following steps: generating current, generating voltage and power limiting instructions of each branch circuit; the operation data of the reference power generating equipment (the same as the operation data of the target power generating equipment) within one month is acquired.

Secondly, the acquired related data can be subjected to data screening. And judging whether corresponding data are reserved or deleted according to the conditions that whether the irradiance exceeds a preset threshold value, whether the electricity limiting instruction controls electricity limiting, and whether the generated current of each branch circuit is in a preset range (for example, the preset range of the current can be 0-15A). And deleting the data of the date when the irradiance does not exceed the preset threshold and the electricity is limited, and deleting the current data of the branch in the date when the generated current is out of the preset range.

The data may then be processed and a determination may be made as to whether short-term occlusion features are present based on the processed data. For the operation data of the target power generation equipment, the date on which the power generation amount is larger than 5 hours and the dispersion rate is lower than 40% can be determined, the effective date is obtained, the minute average value of each group string of the combiner boxes of the effective date is calculated, and for the reference power generation equipment, the minute average value of each group string of the combiner boxes of the reference power generation equipment can be directly calculated. Furthermore, the characteristics of the average value of the minutes of the reference power generation equipment and the target power generation equipment in the same date and the same time period can be compared through a BP neural network classification model, if the characteristics of the average value of the minutes of the group string power generation currents of the reference power generation equipment and the target power generation equipment are different in any time period, the time period is considered to be blocked, and a result is output.

The BP neural network model is trained firstly, the training of the neural network can take an electric field as a unit, and the training for the first time needs to artificially observe data and mark. And with the time interval as a unit, obtaining irradiance, a current mean value, a voltage mean value, the power generation amount of the reference power generation equipment and the power generation amount of the target power generation equipment, and marking the sample data with a label of a positive case (occlusion exists in the time interval) or a negative case (occlusion does not exist in the time interval). Because the generated energy of the photovoltaic power generation equipment is greatly influenced by the sunshine intensity, sunshine time and temperature, the iterative training BP neural network model can be updated again subsequently by using the operation data of the previous 6 months and all historical negative examples as training sets.

The identification model of the short-term occlusion is based on environmental data, SCADA (supervisory control and data acquisition) operation data of photovoltaic power generation equipment and operation data of electric field marker post equipment (reference power generation equipment), the possibility of the short-term occlusion is judged through short-term and time-sharing judgment logics, single prediction is realized through forward propagation of generated energy by adopting a classical back propagation BP (back propagation) neural network, parameters of the BP model can be adjusted through feedback (back propagation) of prediction errors during model training, and the prediction precision is improved.

On the other hand, a specific embodiment of the second submodel is exemplarily described.

The second submodel is used to identify long term occlusions, and the relevant data used herein may specifically include the operational data of the target power generation device.

Specifically, the environmental data may be input to a second sub-model to obtain predicted power generation data of the target power generation device, where the second sub-model may include a long-short term memory LSTM neural network for predicting power generation data, and the power generation data is predicted by the LSTM neural network.

Then, the actual power generation data of the target power generation device may be compared with the predicted power generation data to obtain an occlusion recognition result for long-term occlusion. That is, it is determined whether there is a long-term occlusion according to a comparison of the actual power generation data and the predicted power generation data. This is because when the surface of the power generating equipment is covered with dust, bird droppings, leaves or other sundries, the power generating equipment is in a low power generation and under-power generation state for a long time. Under normal production conditions, the deviation between the actual power generation amount and the regression (predicted) power generation amount of a power generation facility should be within a certain range, and if the deviation exceeds the range, it indicates that the power generation facility has long-term occlusion.

Further, in order to avoid the influence of external factors such as weather on the amount of power generation, the error due to the influence of other factors can be eliminated based on the determination result of the power generation equipment of the entire target electric field. Specifically, the target electric field includes a plurality of target power generation devices, and after actual power generation data and predicted power generation data of each target power generation device are obtained, an error parameter of the actual power generation data and the predicted power generation data of each target power generation device may be calculated, if the number of target power generation devices with the error parameters exceeding the preset error threshold value does not exceed the preset number, determining that the target power generation equipment with the error parameter exceeding the preset error threshold has long-term occlusion, otherwise, if the number of the target power generation devices with the error parameters exceeding the preset error threshold exceeds the preset number, the problem that the power generation amount of a plurality of power generation devices in the target electric field is relatively small is shown, which may not be caused by the long-term shielding of the target power generation devices, other problems may exist, in which case the results may be discarded and new operational data may be retrieved for determination.

FIG. 4 is a flow diagram illustrating an embodiment of identifying long-term occlusions for the second sub-model.

Referring to the second sub-model of fig. 4, the acquired related data includes environmental information (region, terrain, longitude and latitude, etc.) of the target electric field, weather forecast information, SCADA operation data (including generation voltage, generation current, power limit instruction, configuration information of the string), and the like.

Firstly, the weather forecast of the position of the target electric field can be determined based on the information of the region, the terrain, the longitude and latitude and the like of the target electric field and the weather forecast data, so that the data of weather conditions such as rain, snow, cloudy days and the like can be deleted, and the inefficient power generation caused by the weather can be eliminated. And whether power limitation exists can be judged through the operation data, and if no power limitation exists and weather is abnormal, whether long-term shielding exists can be analyzed by using data in the day.

In addition, whether the effective group string in each day is less than 80% of the total group string can be judged according to the group string configuration information of the target power generation equipment, if so, the power generation amount in the day is affected, and the data of the corresponding date is deleted, otherwise, the data of the corresponding date can be reserved, and the actual power in the day can be calculated according to the mean value and the variance of the current and voltage in 5 minutes of each branch circuit.

Next, historical operating data for a plurality of days (e.g., 15 days) may be obtained, and the power generated for the next day may be predicted to obtain predicted power. After the predicted power is obtained, comparing the predicted power and the actual power on the same date, judging whether the MSE (mean square error) deviation is larger than T, if not, judging that the power generation equipment has no long-term occlusion, if so, further judging whether the MSE of most power generation equipment exceeds T or only the MSE of individual equipment exceeds T in the whole field, if only the MSE difference of the individual equipment (for example, less than or equal to 2) in the whole field exceeds T, determining that the corresponding power generation equipment has a long-term occlusion problem, otherwise, if the MSE of more power generation equipment (exceeding 2) exceeds T, possibly due to other reasons, considering that the result is invalid, discarding the judgment result, and selecting an effective date for identifying the long-term occlusion.

Because the power prediction depends on the power of a previous period of time, the selection of the LSTM of the time-cycle neural network can better show the change of the power value of the long-term shielding along with the change of the time, and compared with the traditional cycle neural network, the model operation efficiency and the prediction accuracy are improved.

When the parameters of the LSMT model are adjusted, if the training result has overfitting and the effect is poor in the test data, the input variables may be normalized, for example, by using the L2 regularization method, and the discarding parameters are increased. If the model is under-fitted, this shows that the inexhaustible mining of power data features can remove or reduce regularization while increasing the number of network layers. When the model network is not converged for a long time, the MSE is mainly shown in a larger numerical range for a long time, at the moment, the activation function can be adjusted in the training process, and softsign is adopted to replace the tanh activation function, so that the model training efficiency is improved, and the training saturation is avoided. In addition, for the problem of a model structure or a data set, a network structure can be integrally planned before model training, data is standardized through some common data processing means, and input variables with low contribution rate to power are removed.

Optionally, some optional implementations of the screening and processing of the related data in the examples of the present application are described below.

First, invalid data corresponding to an invalid condition corresponding to a data type may be removed from each data of the data related to the target electric field to obtain valid data, and the valid data may be input to the occlusion recognition model to obtain an occlusion recognition result.

Further, when the operation data includes the branch power generation parameter of each branch per day, the operation data of the branch whose branch power generation parameter is outside the first parameter range may be removed from the operation data of the target power generation device and the operation data of the reference power generation device, respectively.

Further, before the operation data of the branch circuit of which the branch circuit power generation parameter is out of the first parameter range is removed from the operation data of the target power generation equipment, the date on which the target power generation equipment is limited to operate can be determined according to the daily power limiting control instruction, the first date is obtained, and the date on which the daily irradiance is out of the second parameter range is determined, and the second date is obtained; wherein the environmental data of the target electric field comprises daily irradiance; thus, the operation data of the target power generation device on the first date and the second date is deleted; accordingly, in the operation data of the reference power generating equipment, the operation data of the reference power generating equipment on the first date and the second date may also be removed before the operation data of the branch whose branch power generation parameter is out of the first parameter range is removed.

It should be noted that, in the shielding early warning method for the photovoltaic power generation device provided in the embodiment of the present application, the execution main body may be a shielding early warning device for the photovoltaic power generation device, or a control module in the shielding early warning device for the photovoltaic power generation device, which is used for executing the shielding early warning method for the photovoltaic power generation device. In the embodiment of the present application, a method for performing a shielding early warning of a photovoltaic power generation device by using a shielding early warning device of the photovoltaic power generation device is taken as an example, and the shielding early warning device of the photovoltaic power generation device provided in the embodiment of the present application is described.

As shown in fig. 5, a schematic diagram of a blocking early warning device for a photovoltaic power generation apparatus provided in an embodiment of the present application includes a first obtaining unit 51, an input unit 52, and an output unit 53.

The first acquiring unit 51 is configured to acquire relevant data of the target electric field, where the relevant data includes environmental data of the target electric field, operation data of the target power generation equipment in the target electric field, and operation data of the reference power generation equipment in the target electric field;

the input unit 52 is configured to input the relevant data into an occlusion recognition model to obtain an occlusion recognition result, where the occlusion recognition result is used to represent whether the target power generation device is occluded, and the occlusion recognition model includes a first sub-model and a second sub-model, where the first sub-model is used to recognize short-term occlusion and the second sub-model is used to recognize long-term occlusion;

the output unit 53 is configured to output warning information according to the occlusion recognition result, where the warning information is used to indicate whether short-term occlusion and long-term occlusion exist in the target power generation device.

According to the shielding early warning device of the photovoltaic power generation equipment, whether long-term shielding and short-term shielding exist can be determined respectively based on different sub models used for identifying the long-term shielding and the short-term shielding respectively through relevant data such as environment data of a target electric field, operation data of the target power generation equipment in the target electric field, and operation data of reference power generation equipment in the target electric field, and then early warning information is output according to a shielding identification result.

Optionally, the input unit 52 may be further configured to input the operation data of the target power generation device and the operation data of the reference power generation device into the first sub-model, and compare the difference between the target power generation device and the reference power generation device to obtain an occlusion recognition result for the short-term occlusion.

Alternatively, the input unit 52 may include:

Optionally, the first comparing subunit is further configured to determine that short-term occlusion exists in the target power generation device when the first submodel determines that the target power generation device and the reference power generation device are in the same time period exceeding m days and that the branch average power generation parameter is different, where m is a positive integer greater than or equal to 1.

Optionally, the operational data of the target power generation device includes actual power generation data; the input unit 52 may include:

Optionally, the target electric field comprises a plurality of target power generation devices; the second ratio subunit comprises:

Optionally, the apparatus may further include:

the input unit 52 may also be configured to input valid data into the occlusion recognition model, so as to obtain an occlusion recognition result.

Optionally, the operational data comprises branch power generation parameters for each branch per day; the deleting unit is further used for removing the operation data of the branch circuit with the branch circuit power generation parameter out of the first parameter range from the operation data of the target power generation equipment and the operation data of the reference power generation equipment respectively.

Optionally, the apparatus may further include:

the deleting unit may be further configured to remove the operation data of the target power generation device on the first date and the second date, and remove the operation data of the reference power generation device on the first date and the second date before removing the operation data of the branch whose branch power generation parameter is outside the first parameter range, in the operation data of the reference power generation device.

Optionally, the warning information includes a warning level, and the output unit 53 may include:

Optionally, the first sub-model comprises a back propagation BP neural network and the second sub-model comprises a long short term memory LSTM neural network.

Fig. 6 shows a hardware structure diagram of an electronic device provided in an embodiment of the present application.

The electronic device may comprise a processor 301 and a memory 302 in which computer program instructions are stored.

Specifically, the processor 301 may include a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or may be configured to implement one or more Integrated circuits of the embodiments of the present Application.

Memory 302 may include mass storage for data or instructions. By way of example, and not limitation, memory 302 may include a Hard Disk Drive (HDD), floppy Disk Drive, flash memory, optical Disk, magneto-optical Disk, tape, or Universal Serial Bus (USB) Drive or a combination of two or more of these. Memory 302 may include removable or non-removable (or fixed) media, where appropriate. The memory 302 may be internal or external to the integrated gateway disaster recovery device, where appropriate. In a particular embodiment, the memory 302 is a non-volatile solid-state memory.

The memory may include Read Only Memory (ROM), Random Access Memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical, or other physical/tangible memory storage devices. Thus, in general, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software comprising computer-executable instructions and when the software is executed (e.g., by one or more processors), it is operable to perform operations described with reference to the methods according to an aspect of the application.

The processor 301 reads and executes the computer program instructions stored in the memory 302 to implement the occlusion warning method for the photovoltaic power generation apparatus in the above embodiment.

In one example, the electronic device may also include a communication interface 303 and a bus 310. As shown in fig. 6, the processor 301, the memory 302, and the communication interface 303 are connected via a bus 310 to complete communication therebetween.

The communication interface 303 is mainly used for implementing communication between modules, apparatuses, units and/or devices in the embodiment of the present application.

Bus 310 includes hardware, software, or both to couple the components of the online data traffic billing device to each other. By way of example, and not limitation, a bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hypertransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an infiniband interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a video electronics standards association local (VLB) bus, or other suitable bus or a combination of two or more of these. Bus 310 may include one or more buses, where appropriate. Although specific buses are described and shown in the embodiments of the application, any suitable buses or interconnects are contemplated by the application.

It is to be understood that the present application is not limited to the particular arrangements and instrumentality described above and shown in the attached drawings. A detailed description of known methods is omitted herein for the sake of brevity. In the above embodiments, several specific steps are described and shown as examples. However, the method processes of the present application are not limited to the specific steps described and illustrated, and those skilled in the art can make various changes, modifications, and additions or change the order between the steps after comprehending the spirit of the present application.

The functional blocks shown in the above-described structural block diagrams may be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, it may be, for example, an electronic circuit, an Application Specific Integrated Circuit (ASIC), suitable firmware, plug-in, function card, or the like. When implemented in software, the elements of the present application are the programs or code segments used to perform the required tasks. The program or code segments may be stored in a machine-readable medium or transmitted by a data signal carried in a carrier wave over a transmission medium or a communication link. A "machine-readable medium" may include any medium that can store or transfer information. Examples of a machine-readable medium include electronic circuits, semiconductor memory devices, ROM, flash memory, Erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, Radio Frequency (RF) links, and so forth. The code segments may be downloaded via computer networks such as the internet, intranet, etc.

It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, the present application is not limited to the order of the above-described steps, that is, the steps may be performed in the order mentioned in the embodiments, may be performed in an order different from the order in the embodiments, or may be performed simultaneously.

Aspects of the present application are described above with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions/acts specified in the flowchart and/or block diagram block or blocks. Such a processor may be, but is not limited to, a general purpose processor, a special purpose processor, an application specific processor, or a field programmable logic circuit. It will also be understood that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware for performing the specified functions or acts, or combinations of special purpose hardware and computer instructions.

As described above, only the specific embodiments of the present application are provided, and it can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the system, the module and the unit described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again. It should be understood that the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive various equivalent modifications or substitutions within the technical scope of the present application, and these modifications or substitutions should be covered within the scope of the present application.

Claims

1. A shielding early warning method for photovoltaic power generation equipment is characterized by comprising the following steps:

inputting the related data into an occlusion recognition model to obtain an occlusion recognition result, wherein the occlusion recognition result is used for representing whether the target power generation equipment has occlusion, the occlusion recognition model comprises a first sub-model and a second sub-model, the first sub-model is used for recognizing short-term occlusion, and the second sub-model is used for recognizing long-term occlusion;

and outputting early warning information according to the occlusion identification result, wherein the early warning information is used for indicating whether the target power generation equipment has the short-term occlusion and the long-term occlusion.

2. The method of claim 1, wherein inputting the relevant data into an occlusion recognition model to obtain an occlusion recognition result comprises:

and inputting the operation data of the target power generation equipment and the operation data of the reference power generation equipment into the first sub-model, and comparing the difference between the target power generation equipment and the reference power generation equipment to obtain the shielding identification result of the short-term shielding.

3. The method of claim 2, wherein inputting the operational data of the target power generation device and the operational data of the reference power generation device into the first sub-model, comparing the difference between the target power generation device and the reference power generation device to obtain an occlusion recognition result of the short term occlusion, comprises:

according to the operation data of the target power generation equipment and the reference power generation equipment, branch average power generation parameters of the target power generation equipment and the reference power generation equipment in different time periods of each day are respectively calculated;

and comparing the difference of the branch average power generation parameters of the target power generation equipment and the reference power generation equipment in each time period through the first submodel to judge whether the target power generation equipment has the short-term shelter in the corresponding time period.

4. The method according to claim 3, wherein the comparing, by the first sub-model, the difference of the branch average power generation parameters of the target power generation device and the reference power generation device in each time period to determine whether the target power generation device has the short-term occlusion in the corresponding time period comprises:

and determining that the target power generation equipment has the short-term shelter when the first sub-model judges that the target power generation equipment and the reference power generation equipment have difference in the average branch power generation parameters in the same time period exceeding m days, wherein m is a positive integer greater than or equal to 1.

5. The method of claim 1, wherein the operational data of the target power generation device comprises actual power generation data;

the step of inputting the relevant data into an occlusion recognition model to obtain an occlusion recognition result includes:

inputting the environmental data into the second submodel to obtain predicted power generation data of the target power generation equipment;

6. The method of claim 5, wherein a plurality of said target power generation devices are included in said target electric field;

the comparing the actual power generation data of the target power generation device with the predicted power generation data to obtain a result of occlusion recognition of the long-term occlusion includes:

calculating an error parameter of the actual power generation data and the predicted power generation data of each of the target power generation devices;

and under the condition that the number of the target power generation devices with the error parameters exceeding the preset error threshold does not exceed the preset number, determining that the target power generation devices with the error parameters exceeding the preset error threshold have the long-term shielding.

7. The method of claim 1, wherein after the obtaining data related to the electric field of the target, further comprising:

respectively removing invalid data which are consistent with invalid conditions corresponding to the data types from each data of the related data of the target electric field to obtain valid data;

the step of inputting the relevant data into an occlusion recognition model to obtain an occlusion recognition result includes: and inputting the effective data into the occlusion recognition model to obtain the occlusion recognition result.

8. The method of claim 7, wherein the operational data includes branch power generation parameters for each branch per day;

the removing of the invalid data corresponding to the invalid condition corresponding to the data type from each data of the data related to the target electric field, respectively, includes:

9. The method according to claim 8, wherein the removing operation data of the branch circuit in which the branch circuit power generation parameter is outside the first parameter range from the operation data of the target power generation device further comprises:

determining the date of which the daily irradiance is out of the second parameter range to obtain a second date; wherein the environmental data of the target electric field comprises the daily irradiance;

before removing the operation data of the branch circuit of which the branch circuit power generation parameter is out of the first parameter range from the operation data of the reference power generation equipment, the method further comprises the following steps:

removing operating data of the reference power plant on the first date and the second date.

10. The method according to any one of claims 1-9, wherein the pre-warning information comprises a pre-warning level, and the outputting pre-warning information according to the occlusion recognition result comprises:

responding to the occlusion identification result that the short-term occlusion exists, and outputting the early warning level as a first level and the time period of the short-term occlusion;

responding to the occlusion identification result that the long-term occlusion exists, and outputting the early warning grade as a second grade;

responding to the occlusion identification result that the short-term occlusion and the long-term occlusion exist at the same time, and outputting the early warning level as a third level and a time period of the short-term occlusion;

wherein the levels of the first level, the second level and the third level are from low to high.

11. The method of any of claims 1-9, wherein the first sub-model comprises a back-propagation (BP) neural network and the second sub-model comprises a long-short-term memory (LSTM) neural network.

12. The utility model provides a photovoltaic power generation equipment shelter from early warning device which characterized in that includes:

the input unit is used for inputting the related data into an occlusion recognition model to obtain an occlusion recognition result, the occlusion recognition result is used for representing whether the target power generation equipment is occluded or not, the occlusion recognition model comprises a first submodel and a second submodel, the first submodel is used for recognizing short-term occlusion, and the second submodel is used for recognizing long-term occlusion;

and the output unit is used for outputting early warning information according to the shielding identification result, wherein the early warning information is used for indicating whether the target power generation equipment has the short-term shielding and the long-term shielding.

13. The apparatus according to claim 11, wherein the input unit is further configured to input the operation data of the target power generation device and the operation data of the reference power generation device into the first sub-model, and compare the difference between the target power generation device and the reference power generation device to obtain the occlusion recognition result of the short-term occlusion.

14. The apparatus of claim 13, wherein the input unit comprises:

and the first comparison subunit is used for comparing the difference of the branch average power generation parameters of the target power generation equipment and the reference power generation equipment in each time period through the first submodel to judge whether the target power generation equipment has the short-term shelter in the corresponding time period.

15. The apparatus of claim 14, wherein the first comparison subunit is further configured to determine that the short term occlusion exists for the target power generation device if the first submodel determines that the target power generation device and the reference power generation device have a difference in the branch average power generation parameters over a same period of time exceeding m days, where m is a positive integer greater than or equal to 1.

16. The apparatus of claim 12, wherein the operational data of the target power generation device comprises actual power generation data;

the input unit includes:

the input sub-unit is used for inputting the environment data into the second sub-model to obtain predicted power generation data of the target power generation equipment;

17. The apparatus of claim 16, wherein a plurality of said target power generation devices are included in said target electric field;

the second ratio subunit comprises:

a second calculation subunit configured to calculate an error parameter of the actual power generation data and the predicted power generation data of each of the target power generation devices;

the determining subunit is configured to determine that the target power generation device with the error parameter exceeding the preset error threshold has the long-term occlusion when the number of the target power generation devices with the error parameter exceeding the preset error threshold does not exceed the preset number.

18. The apparatus of claim 12, further comprising:

the deleting unit is used for removing invalid data which are consistent with invalid conditions corresponding to the data types from each type of data of the related data of the target electric field respectively after the related data of the target electric field are obtained to obtain valid data;

the input unit is further configured to input the valid data into the occlusion recognition model to obtain the occlusion recognition result.

19. The apparatus of claim 18, wherein the operational data comprises branch power generation parameters for each branch per day;

the deleting unit is further configured to remove, from the operation data of the target power generation device and the operation data of the reference power generation device, the operation data of the branch whose branch power generation parameter is outside the first parameter range, respectively.

20. The apparatus of claim 19, further comprising:

the first determining unit is used for determining the date of the target power generation equipment which is limited to operate according to the daily power limiting control instruction before the operation data of the branch circuit of which the branch circuit power generation parameter is out of the first parameter range is removed from the operation data of the target power generation equipment to obtain a first date;

the second determining unit is used for determining the date of which the daily irradiance is out of the second parameter range to obtain a second date; wherein the environmental data of the target electric field comprises the daily irradiance;

the deleting unit is further configured to remove the operation data of the target power generation device on the first date and the second date, and remove the operation data of the reference power generation device on the first date and the second date before removing the operation data of the branch whose branch power generation parameter is outside the first parameter range from the operation data of the reference power generation device.

21. The apparatus of any one of claims 12-20, wherein the warning information includes a warning level, and the output unit includes:

the first output subunit is used for responding to the short-term occlusion existence of the occlusion identification result, and outputting the early warning level as a first level and the time period of the short-term occlusion;

the second output subunit is used for responding to the occlusion identification result that the long-term occlusion exists and outputting the early warning grade as a second grade;

a third output subunit, configured to output, in response to the occlusion recognition result indicating that the short-term occlusion and the long-term occlusion exist at the same time, a third-level early warning level and a period of the short-term occlusion;

22. The apparatus of any of claims 12-20, wherein the first sub-model comprises a back-propagation BP neural network and the second sub-model comprises a long-short-term memory, LSTM, neural network.

23. An electronic device, characterized in that the electronic device comprises: a processor and a memory storing computer program instructions;

the processor, when executing the computer program instructions, implements the occlusion warning method of a photovoltaic power generation apparatus as claimed in any one of claims 1-11.

24. A storage medium, characterized in that the storage medium has stored thereon computer program instructions which, when executed by a processor, implement the occlusion warning method of a photovoltaic power generation apparatus according to any one of claims 1 to 11.