Personal photovoltaic power generation filing data verification method and system

Abstract

The invention provides a personal photovoltaic power generation filing data verification method and system, and the method comprises the steps: arranging and screening filing data when a user installs photovoltaic power generation, and obtaining preprocessed filing data; the latest high-resolution remote sensing image of the area where the user is located is analyzed, accurate measurement of the area of the photovoltaic panel is achieved, and therefore whether the preprocessed record data is correct or not is verified; analyzing the power consumption behavior characteristics before and after the user installs the photovoltaic power generation panel, if the power consumption after photovoltaic installation is analyzed to be increased, judging that the record data is correct, otherwise, judging that the record data is wrong; judging the power consumption behavior after the energy storage electric quantity of the user is used up, if the power consumption during photovoltaic power generation is far less than the power consumption during power supply of the power grid, marking that the record data is wrong, otherwise, marking that the record data is correct; and comprehensively analyzing whether the photovoltaic power generation filing data are wrong or not according to the marking result obtained in the previous step. According to the invention, verification of personal photovoltaic power generation filing data can be comprehensively realized.

Description

Technical Field

[0001] The present invention relates to the use of satellite remote sensing image data and user electricity consumption behavior analysis to realize the verification of photovoltaic power generation registration data, and specifically to a personal photovoltaic power generation registration data verification method and system. Background Art

[0002] In recent years, with the development of photovoltaic technology and energy storage technology, photovoltaic power generation, as a representative of green energy, has received widespread attention. With the gradual reduction of costs and the steady improvement of efficiency, more and more individuals are gradually accepting photovoltaic power generation equipment as an effective supplement to power grid distribution. On the other hand, the installation and use of photovoltaic equipment need to be managed in an orderly manner under safe and stable conditions. Individuals using photovoltaic equipment need to register with the power department: on the one hand, it can optimize the distribution method of traditional electricity, and at the same time, it also puts the use and maintenance of photovoltaic equipment under long-term safety supervision.

[0003] However, after the initial filing, individuals who use photovoltaic equipment often complete the subsequent capacity expansion and energy storage upgrade privately without going to the relevant departments for secondary filing. This has a certain impact on the power department's effective approval of regional power loads and optimization of power distribution networks; more importantly, private installation, expansion, and stamping of photovoltaic equipment without strict quality control poses a greater safety risk. Summary of the invention

[0004] The present invention mainly aims at the accuracy of personal photovoltaic power generation registration data, and provides a personal photovoltaic power generation registration data verification method to achieve the measurement of photovoltaic panel area through satellite remote sensing data; then compare and analyze the data before and after the user installs the photovoltaic panel, as well as the electricity consumption behavior under extreme weather conditions, to comprehensively realize the verification of personal photovoltaic power generation registration data.

[0005] The above technical problems of the present invention are mainly solved by the following technical solutions:

[0006] A method for verifying personal photovoltaic power generation filing data, comprising the following steps:

[0007] Step 1: Preprocessing of photovoltaic power generation filing data: sort out and screen the filing data of users when installing photovoltaic power generation, and record the user's address, installed capacity, energy storage capacity, deployment location, photovoltaic panel installation angle and actual area of ​​photovoltaic panels as preprocessed filing data;

[0008] Step 2: Verify the area of ​​photovoltaic panels based on satellite remote sensing images: Analyze the latest high-resolution remote sensing images of the user's area to accurately measure the area of ​​photovoltaic panels, so as to verify whether the pre-processed filing data obtained in step 1 is correct;

[0009] Step 3: Compare and analyze the electricity consumption behavior before and after the photovoltaic installation: Analyze the electricity consumption characteristics of the user before and after the installation of the photovoltaic panels. If the analysis shows that the electricity consumption increases after the photovoltaic installation, it is judged that the "recorded data is correct", otherwise it is judged that the "recorded data is incorrect";

[0010] Step 4: Analysis of electricity consumption behavior in "extreme weather": Determine the user's electricity consumption behavior after the energy storage power is used up. If the power consumption during photovoltaic power generation is much less than the power consumption during grid power supply, it is marked as "the registered data is incorrect", otherwise it is marked as "the registered data is correct";

[0011] Step 5: Photovoltaic power generation registration data verification: Based on the marking results obtained in steps 2, 3 and 4, comprehensively analyze whether the photovoltaic power generation registration data is incorrect.

[0012] Furthermore, step one specifically includes:

[0013] Step 1.1: Collect the filing data submitted by the user when installing the photovoltaic power generation system, and extract 5 types of data including user address, installed capacity, energy storage capacity, deployment location and photovoltaic panel installation inclination;

[0014] Step 1.2: Convert the installed capacity, photovoltaic type and brand into the actual area of ​​photovoltaic panels and record it;

[0015] Step 1.3: According to the relevant national standards and industry standards, the collected filing data and the converted actual area of ​​PV panels are screened to select the data that meets the standards;

[0016] Step 1.4: After the final screening is qualified, the six data including user address, installed capacity, energy storage capacity, deployment location, photovoltaic panel installation angle and actual area of ​​photovoltaic panels will be recorded as pre-processed filing data.

[0017] Furthermore, step 2 specifically includes:

[0018] Step 2.1, collect multiple high-resolution remote sensing images of the user's area, with a resolution greater than 1 meter;

[0019] Step 2.2: Convert the user address provided in the user filing data into longitude and latitude coordinates;

[0020] Step 2.3, mark the longitude and latitude coordinates in the high-resolution remote sensing image;

[0021] Step 2.4: Based on the actual area of ​​the photovoltaic panel obtained in step 1, record it as Area, and circle the area 10 times the actual area of ​​the photovoltaic panel around the positioning coordinates in the high-resolution image, that is, observe the image area of ​​10Area;

[0022] Step 2.5: For the collected high-resolution remote sensing images, intercept the images of 10Area areas to obtain n images, which are recorded as IM n ;

[0023] Step 2.6, extracting sensor parameter data of the image, wherein the sensor parameter data includes image shooting time and image shooting angle;

[0024] Step 2.7: Analyze the position and angle relationship of the sun, photovoltaic panel, and satellite according to the image shooting time and image shooting angle: if the satellite is within the reflection angle of sunlight, the image shot at that time point will be eliminated;

[0025] Step 2.8: In the remaining remote sensing images, build a deep learning model to extract the visual area of ​​the photovoltaic panels;

[0026] Step 2.9: Convert the visual area into the actual area of ​​the photovoltaic panel according to the photovoltaic panel installation angle and the image shooting angle;

[0027] Step 2.10, record the maximum value of the actual area of ​​the photovoltaic panel calculated from multiple images, and record it as realArea;

[0028] Step 2.11: Compare the Area and realArea values. If the difference between the two is less than 5%, it is marked as "the registered data is correct".

[0029] Furthermore, step 2 also includes:

[0030] Step 2.12: If the difference between Area and realArea is greater than 5%, it is considered that other registered photovoltaic panels may be included in the statistics, and further processing is required:

[0031] a) First, count whether there are other photovoltaic power generation records within a certain range of photovoltaic record coordinates. If not, remove other record interferences and define it as "wrong record data";

[0032] b) If there are other registered data within a certain range, repeat steps 2.4-2.5 for the other registered data to extract 10 times the area of ​​all registered photovoltaic panel images within the range;

[0033] c) For the realArea selected in step 2.10, backtrack to find the corresponding high-resolution remote sensing image;

[0034] d) extracting the photovoltaic panel areas of all registered users in the region based on the remote sensing high-resolution image found in step c);

[0035] e) Compare the calculated photovoltaic panel areas of all registered users with the sum of the registered data. If the error is less than 5%, the initial user's "registered data is correct" is marked;

[0036] f) If not, mark it as “the registered data is incorrect”.

[0037] Furthermore, step three specifically includes:

[0038] Step 3.1, statistics the user's electricity consumption behavior in the year before the installation of the photovoltaic power generation system, including electricity consumption time period and electricity consumption;

[0039] Step 3.2: Collect and compile weather information for each time period in the year before the installation of photovoltaic power generation, including temperature, humidity, air pressure, rainfall, and snowfall;

[0040] Step 3.3: Build a random forest model, and use the various data from the previous year to make a sample set for training;

[0041] Step 3.4: Count the weather and electricity consumption characteristics in different time periods after the installation of photovoltaic panels;

[0042] Step 3.5: Calculate the daily power generation after installing the photovoltaic panels based on the registered photovoltaic system data and weather data;

[0043] Step 3.6: Assume that the installed capacity of photovoltaic power generation of a certain user is P d At the beginning of the peak power consumption period of the day, the remaining energy storage capacity P s , assuming that the energy storage capacity is 15% of the photovoltaic installed capacity, then P s <0.15P d ;

[0044] Step 3.7: Use the random forest model in step 3.3 to predict the electricity consumption on all dates and in all time periods after the installation of the photovoltaic panels; let the predicted electricity consumption of the user in a certain time period T be P h ;

[0045] Step 3.6: Assume that the power consumption of the user in the T period is P e , photovoltaic power consumption is P p , then P p <P s ;

[0046] Step 3.7, then P e +P p is the actual power consumption of the user, P e +P p <P e +0.15P d ;

[0047] Step 3.8: Compare P h and Pe +0.15P d , if P h <P e +0.15P d , it indicates that the registered photovoltaic power generation is redundant, or the power consumption increases after the user installs photovoltaic power generation; this situation is marked as "registered data is correct";

[0048] Step 3.9: If P h Much larger than P e +0.15P d , such as P h >1.5(P e +0.15P d ), this situation shows that after the user installed photovoltaic power generation, the electricity consumption dropped sharply, which was marked as "the registered data is wrong", where 1.5 is the empirical coefficient, indicating that the former is 1.5 times greater than the latter, which is used to quantify "far greater than".

[0049] Furthermore, step four specifically includes:

[0050] Step 4.1: If a day is defined as "extreme weather", analyze the electricity consumption data of users on that day;

[0051] Step 4.2: If the user does not use grid power distribution, it means that the user is still using energy storage power;

[0052] Step 4.3: Assume that the extreme weather lasts for n days in total, and further analyze the continuous electricity consumption characteristics of users on day n+1:

[0053] c) Assume that in the n+1 days, the first m days use photovoltaic power supply, the kth day uses hybrid power supply, and the next j days use grid distribution, where k = m+1;

[0054] b) Assume that the average power consumption in the previous m days is P m , the energy storage power P used on the kth day k1 ;

[0055] c) The average actual power consumption in the next j days can be monitored and is set as P j ;

[0056] d) The power distribution consumption of the power grid on the kth day can also be obtained, set as P k2 ;

[0057] e) Get mP m +P k1 <0.15P d ;

[0058] f) Assuming that the power consumption on the kth day is close to the average power consumption on the next j days, estimate P k1 Equal to P j -Pk2 ;

[0059] g) Estimate the daily energy storage power consumption in the previous m days:

[0060]

[0061] h) Compare P m and P j Size relationship: If Pm>P j , indicating that the power consumption during photovoltaic power generation is large, and marked as "the registered data is correct";

[0062] iii) If P m Much smaller than P j , such as 1.5P m <P j , it means that the data of photovoltaic power generation is far less than the power consumption of grid power supply, and it is marked as "the registered data is wrong".

[0063] Furthermore, step five specifically includes:

[0064] Step 5.1: If steps 2, 3, and 4 are all marked as "incorrect filing data", the final definition is "incorrect filing data", and staff are required to conduct on-site verification;

[0065] Step 5.2: If any two of the steps 2, 3, and 4 are marked as "incorrect data on record", you will need to contact us by phone to verify and confirm based on the latest high-resolution image visual interpretation;

[0066] Step 5.3: If any of the steps 2, 3, and 4 is marked as "incorrect registration data", mark the user registration status as "pending confirmation" and repeat steps 1 to 4 after a certain period for analysis and verification.

[0067] A personal photovoltaic power generation registration data verification system, comprising:

[0068] Photovoltaic power generation filing data preprocessing module, used to sort out and filter the filing data of users when installing photovoltaic power generation, and record the user's address, installed capacity, energy storage capacity, deployment location, photovoltaic panel installation angle and actual area of ​​photovoltaic panels as preprocessed filing data;

[0069] The photovoltaic panel area verification module based on satellite remote sensing images is used to analyze the latest high-resolution remote sensing images of the user's area to achieve accurate measurement of the photovoltaic panel area, so as to verify whether the pre-processed filing data obtained in step 1 is correct;

[0070] The module for comparing and analyzing electricity consumption behavior before and after photovoltaic installation is used to analyze the electricity consumption characteristics of users before and after installing photovoltaic panels. If the analysis shows that the electricity consumption increases after photovoltaic installation, it is judged that "the registered data is correct", otherwise it is judged that "the registered data is incorrect";

[0071] The "extreme weather" electricity consumption behavior analysis module is used to determine the user's electricity consumption behavior after the energy storage power is used up. If the power consumption during photovoltaic power generation is much less than the power consumption during grid power supply, it is marked as "the registered data is incorrect", otherwise it is marked as "the registered data is correct";

[0072] The photovoltaic power generation registration data verification module is used to comprehensively analyze whether the photovoltaic power generation registration data is incorrect based on the marking results obtained by the photovoltaic panel area verification module based on satellite remote sensing images, the electricity consumption behavior comparison analysis module before and after photovoltaic installation, and the "extreme weather" electricity consumption behavior analysis module.

[0073] Furthermore, the power consumption behavior comparison and analysis module before and after photovoltaic installation is specifically used for:

[0074] Step 3.1, statistics the user's electricity consumption behavior in the year before the installation of the photovoltaic power generation system, including electricity consumption time period and electricity consumption;

[0075] Step 3.2: Collect and compile weather information for each time period in the year before the installation of photovoltaic power generation, including temperature, humidity, air pressure, rainfall, and snowfall;

[0076] Step 3.3: Build a random forest model, and use the various data from the previous year to make a sample set for training;

[0077] Step 3.4: Count the weather and electricity consumption characteristics in different time periods after the installation of photovoltaic panels;

[0078] Step 3.5: Calculate the daily power generation after installing the photovoltaic panels based on the registered photovoltaic system data and weather data;

[0079] Step 3.6: Assume that the installed capacity of photovoltaic power generation of a certain user is P d At the beginning of the peak power consumption period of the day, the remaining energy storage capacity P s , assuming that the energy storage capacity is 15% of the photovoltaic installed capacity, then P s <0.15P d ;

[0080] Step 3.7: Use the random forest model in step 3.3 to predict the electricity consumption on all dates and in all time periods after the installation of the photovoltaic panels; let the predicted electricity consumption of the user in a certain time period T be P h ;

[0081] Step 3.6: Assume that the power consumption of the user in the T period is P e , photovoltaic power consumption is Pp , then P p <P s ;

[0082] Step 3.7, then P e +P p is the actual power consumption of the user, P e +P p <P e +0.15P d ;

[0083] Step 3.8: Compare P h and P e +0.15P d , if P h <P e +0.15P d , it indicates that the registered photovoltaic power generation is redundant, or the power consumption increases after the user installs photovoltaic power generation; this situation is marked as "registered data is correct";

[0084] Step 3.9: If P h Much larger than P e +0.15P d , such as P h >1.5(P e +0.15P d ), this situation shows that after the user installed photovoltaic power generation, the electricity consumption dropped sharply, which was marked as "the registered data is wrong", where 1.5 is the empirical coefficient, indicating that the former is 1.5 times greater than the latter, which is used to quantify "far greater than".

[0085] Furthermore, the "extreme weather" electricity consumption behavior analysis module is specifically used to:

[0086] Step 4.1: If a day is defined as "extreme weather", analyze the electricity consumption data of users on that day;

[0087] Step 4.2: If the user does not use grid power distribution, it means that the user is still using energy storage power;

[0088] Step 4.3: Assume that the extreme weather lasts for n days in total, and further analyze the continuous electricity consumption characteristics of users on day n+1:

[0089] d) Assume that in the n+1 days, the first m days use photovoltaic power supply, the kth day uses hybrid power supply, and the next j days use grid power distribution, where k = m+1;

[0090] b) Assume that the average power consumption in the previous m days is P m , the energy storage power P used on the kth day k1 ;

[0091] c) The average actual power consumption in the next j days can be monitored and is set as P j ;

[0092] d) The power distribution consumption of the power grid on the kth day can also be obtained, set as P k2 ;

[0093] e) Get mP m +P k1 <0.15P d ;

[0094] f) Assuming that the power consumption on the kth day is close to the average power consumption on the next j days, estimate P k1 Equal to P j -P k2 ;

[0095] g) Estimate the daily energy storage power consumption in the previous m days:

[0096]

[0097] h) Compare P m and P j Size relationship: If Pm>P j , indicating that the power consumption during photovoltaic power generation is large, and marked as "the registered data is correct";

[0098] iv) If P m Much smaller than P j , such as 1.5P m <P j , it means that the data of photovoltaic power generation is far less than the power consumption of grid power supply, and it is marked as "the registered data is wrong".

[0099] The present invention has the following beneficial effects:

[0100] 1. A complete set of inspection methods and systems have been built to intelligently detect whether photovoltaic filing data is accurate;

[0101] 2. By identifying the area of ​​photovoltaic panels in high-resolution remote sensing images, it is used to quickly detect whether the registered data is accurate. At the same time, the angular relationship between the sun, photovoltaic panels, and shooting satellites is considered (if the shooting angle is within the range of sunlight reflection, it is defaulted to mirror reflection and the data is not adopted at this time), so as to maximize the accurate estimation of the photovoltaic panel area;

[0102] 3. The efficiency of photovoltaic energy storage equipment is fully considered, and the accuracy of the photovoltaic installation record data is further evaluated by comparing the user's electricity consumption characteristics before and after photovoltaic installation: in the same time period and under the same weather conditions, the total electricity consumption of users after photovoltaic installation is greatly reduced, which indicates that there may be problems with the record data;

[0103] 4. By defining "extreme weather", we can accurately evaluate the electricity consumption behavior of users when the photovoltaic energy storage is exhausted, and compare and quantify the electricity consumption behavior when the power grid is supplying electricity, further realizing the accuracy evaluation of the registered data. BRIEF DESCRIPTION OF THE DRAWINGS

[0104] Figure 1 It is a flow chart of a method for verifying personal photovoltaic power generation filing data according to an embodiment of the present invention;

[0105] Figure 2 This is a module block diagram of a personal photovoltaic power generation filing data verification system in this embodiment;

[0106] Figure 3 It is a schematic diagram of remote sensing image data screening of the present invention. DETAILED DESCRIPTION

[0107] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. The entire system is implemented in software.

[0108] See also Figure 1 The embodiment of the present invention provides a method for verifying personal photovoltaic power generation registration data, comprising the following steps:

[0109] Step 1: Preprocessing of photovoltaic power generation record data, which is used to organize the record data when the user installs photovoltaic power generation. The specific implementation steps are as follows:

[0110] Step 1.1: Collect the filing data submitted by the user when installing the photovoltaic power generation system, and extract 5 types of data including user address, installed capacity, energy storage capacity, deployment location and photovoltaic panel installation inclination;

[0111] Step 1.2: Convert the installed capacity, photovoltaic type and brand into the actual area of ​​photovoltaic panels and record it;

[0112] Step 1.3: According to the relevant national standards and industry standards, the collected filing data and the converted actual area of ​​PV panels are screened to select data that meets the standards, such as energy storage capacity not less than 15% of the installed capacity and not more than 100%;

[0113] Step 1.4: After the final screening is qualified, the six data including user address, installed capacity, energy storage capacity, deployment location, photovoltaic panel installation angle and actual area of ​​photovoltaic panels will be recorded as pre-processed filing data.

[0114] Step 2: Check the area of ​​photovoltaic panels based on satellite remote sensing images. Since photovoltaic panels need to be deployed in an environment without sunlight blocking, there is no need to consider the situation where vegetation or buildings block the sunlight all the time. This step is used to analyze the latest high-resolution remote sensing images of the user's area to accurately measure the area of ​​photovoltaic panels, so as to verify whether the pre-processed filing data obtained in step 1 is correct. The specific implementation steps are as follows:

[0115] Step 2.1: Collect multiple high-resolution remote sensing images of the user's area, with a resolution greater than 1 meter (such as GF-2 satellite);

[0116] Step 2.2: Convert the user address provided in the user filing data into longitude and latitude coordinates;

[0117] Step 2.3, mark the longitude and latitude coordinates in the high-resolution remote sensing image;

[0118] Step 2.4: Based on the actual area of ​​the photovoltaic panel obtained in step 1 (denoted as Area), circle an area 10 times the actual area of ​​the photovoltaic panel around the positioning coordinates in the high-resolution image, that is, observe the image area of ​​10Area;

[0119] Step 2.5: For the collected high-resolution remote sensing images, intercept the images of 10Area areas to obtain n images, which are recorded as IM n ;

[0120] Step 2.6, extracting sensor parameter data of the image, wherein the sensor parameter data includes image shooting time and image shooting angle (satellite position);

[0121] Step 2.7: Analyze the position and angle relationship of the sun, photovoltaic panel, and satellite based on the image shooting time and image shooting angle: If the satellite is within the reflection angle of sunlight, such as Figure 3 In the interval marked in , the photovoltaic panels reflecting sunlight may cause inaccurate remote sensing data, so the images taken at this time point are removed;

[0122] Step 2.8: In the remaining remote sensing images, build a deep learning model to extract the visual area of ​​the photovoltaic panels (the area in the above figure);

[0123] Step 2.9: Convert the visual area into the actual area of ​​the photovoltaic panel according to the photovoltaic panel installation angle and the image shooting angle;

[0124] Step 2.10, record the maximum value of the actual area of ​​the photovoltaic panel calculated from multiple images, and record it as realArea;

[0125] Step 2.11, compare the Area and realArea values. If the difference between the two is less than 5%, it is marked as "the registered data is correct";

[0126] Step 2.12: If the difference between the two is greater than 5%, it is considered that the photovoltaic panels registered by others (such as neighbors) may be included in the statistics, and further processing is required:

[0127] a) First, count whether there are other photovoltaic power generation registrations within a certain range of the photovoltaic registration coordinates (such as a radius of 100 meters). If not, eliminate the interference of other registrations and define it as "incorrect registration data"; here we mainly observe whether there are other photovoltaic panel registrations (more than one) within a radius of 100 meters. If there are multiple photovoltaic panels registered in adjacent areas (even right next to each other), the photovoltaic panels seen from the high-resolution remote sensing effect may not belong to the same company, which may interfere with the judgment of the photovoltaic panel area of ​​one of them.

[0128] b) If there are other registered data within a certain range, repeat steps 2.4-2.5 for the other registered data to extract 10 times the area of ​​all registered photovoltaic panel images within the range;

[0129] c) For the realArea selected in step 2.10, backtrack to find the corresponding high-resolution remote sensing image;

[0130] d) Considering that the orientation of photovoltaic panels is usually the same within a small area, the photovoltaic panel areas of all registered users in the area are extracted based on the high-resolution image found in step c);

[0131] e) Compare the calculated photovoltaic panel areas of all registered users with the sum of the registered data. If the error is less than 5%, the initial user's "registered data is correct" is marked;

[0132] f) If not, mark it as “the registered data is incorrect”.

[0133] Step 3: Comparative analysis of electricity consumption before and after photovoltaic installation, used to analyze the electricity consumption characteristics of users before and after installing photovoltaic panels: if the analysis shows that the electricity consumption increases after photovoltaic installation, it is judged that "the registered data is correct", otherwise it is judged that "the registered data is incorrect". The specific implementation steps are:

[0134] Step 3.1, statistics the user's electricity consumption behavior in the year before the installation of the photovoltaic power generation system, including electricity consumption time period and electricity consumption;

[0135] Step 3.2: Collect and compile weather information for each time period in the year before the installation of photovoltaic power generation, including temperature, humidity, air pressure, rainfall, and snowfall;

[0136] Step 3.3: Build a random forest model, and use the data from various time periods in the previous year to create a sample set for training;

[0137] Step 3.4: In the same way, the weather and electricity consumption characteristics of different time periods after the installation of photovoltaic panels are counted;

[0138] Step 3.5: Calculate the daily power generation after installing the photovoltaic panels based on the registered photovoltaic system data and weather data (including sunshine duration);

[0139] Step 3.6: Since photovoltaic power generation occurs during the day, and the nighttime is the peak period for individual users, it is necessary to combine energy storage capacity (usually 15%-50% of photovoltaic installed capacity) to comprehensively analyze the electricity consumption behavior after the installation of photovoltaic panels; suppose the photovoltaic installed capacity of a certain user is P d At the beginning of the peak power consumption period (before 18:00 at night), the remaining energy storage capacity P s Assuming that the energy storage capacity is 15% of the photovoltaic installed capacity, it can be seen that P s <0.15P d ;

[0140] Step 3.7: Use the random forest model in step 3.3 (training data before installing photovoltaic panels) to predict the power consumption on all dates and in all time periods after installing photovoltaic panels; suppose the user is in a certain time period T (18:00-24:00 at night), and the predicted power consumption is P h ;

[0141] Step 3.6: Assume that the power consumption of the user in the T period is P e (can be queried), photovoltaic power consumption is P p (cannot be queried), then we know that P p <P s ;

[0142] Step 3.7, then P e +P p is the actual power consumption of the user, P e +P p <P e +0.15P d ;

[0143] Step 3.8: Compare P h and P e +0.15P d , if P h <P e +0.15P d , it indicates that the registered photovoltaic power generation is redundant, or the power consumption increases after the user installs photovoltaic power generation; this situation is marked as "registered data is correct";

[0144] Step 3.9: If Ph Much larger than P e +0.15P d , such as P h >1.5(P e +0.15P d ), this situation shows that after the user installed photovoltaic power generation, the electricity consumption dropped sharply, which was marked as "the registered data is wrong", where 1.5 is the empirical coefficient, indicating that the former is 1.5 times greater than the latter, which is used to quantify "far greater than".

[0145] Step 4: "Extreme weather" electricity consumption behavior analysis step. This step is used to further determine the user's electricity consumption behavior after the energy storage power is used up: if the electricity consumption during photovoltaic power generation is much less than the electricity consumption during power supply by the grid, it is marked as "the registered data is incorrect", otherwise it is marked as "the registered data is correct". Among them, "extreme weather" is defined as: the weather was good the day before, the photovoltaic equipment was generating electricity normally, and the user had electricity consumption behavior, and based on the analysis of electricity consumption behavior, the electricity consumption of the previous day was predicted. If rain and snow occurred on the current day (determined to be poor photovoltaic power generation effect), it would be "extreme weather"; the specific implementation steps are:

[0146] Step 4.1: If a day is defined as "extreme weather", analyze the electricity consumption data of users on that day;

[0147] Step 4.2: If the user does not use grid power distribution, it means that the user is still using energy storage power;

[0148] Step 4.3: Assuming that the extreme weather lasts for n days in total, the continuous electricity consumption characteristics of users on n+1 days (including the night before the extreme weather) can be further analyzed:

[0149] e) Assume that in the n+1 days, the first m days use photovoltaic power supply, the kth day (k=m+1) uses hybrid power supply, and the next j days use grid power distribution;

[0150] b) Assume that the average power consumption in the previous m days is P m , the energy storage power P used on the kth day k1 ;

[0151] c) The average actual power consumption in the next j days can be monitored and set as P j ;

[0152] d) The power distribution consumption of the power grid on the kth day can also be obtained, set as P k2 ;

[0153] e) Then we know that mP m +P k1 <0.15P d ;

[0154] f) Assuming that the power consumption on the kth day is close to the average power consumption on the next j days, P can be estimated. k1 Approximately equal to P j -P k2 ;

[0155] g) The daily energy storage power consumption in the previous m days can be estimated:

[0156]

[0157] h) Compare P m and P j Size relationship: If Pm>P j , indicating that the power consumption during photovoltaic power generation is large, and marked as "the registered data is correct";

[0158] v) If P m Much smaller than P j , such as 1.5P m <P j , which means that the data of photovoltaic power generation is much smaller than

[0159] The amount of electricity used when the power grid is supplying electricity is marked as "the registered data is incorrect".

[0160] As explained, step three is: comparing photovoltaic power consumption with forecast data (because actual power consumption is calculated by analogy); step four: assuming that extreme weather will exhaust the energy storage capacity, the power consumption under similar weather conditions can be calculated through past data for analogy.

[0161] Step 5: Photovoltaic power generation record data verification step: Based on the marking results obtained in the previous steps, comprehensively analyze whether the photovoltaic power generation record data is incorrect. The specific implementation steps are:

[0162] Step 5.1: If steps 2, 3, and 4 are all marked as "incorrect filing data", the final definition is "incorrect filing data", and staff are required to conduct on-site verification;

[0163] Step 5.2: If any two of the steps 2, 3, and 4 are marked as "incorrect data on record", you will need to contact us by phone to verify and confirm based on the latest high-resolution image visual interpretation;

[0164] Step 5.3: If any of the steps 2, 3, and 4 is marked as "incorrect registration data", the user registration status can be marked as "pending" and all the above steps can be repeated after a certain period of time for analysis and verification.

[0165] like Figure 2As shown, an embodiment of the present invention also provides a personal photovoltaic power generation registration data verification system. Taking the photovoltaic power generation facilities installed and registered by a user in a certain area of ​​Wuhan City as an example, the registration data is: the photovoltaic installed capacity is 3kw, the energy storage capacity is 0.45kw, the photovoltaic panel faces due east, the inclination angle is 15 degrees, and the photovoltaic panel area is 18 square meters.

[0166] The system comprises:

[0167] Photovoltaic power generation filing data preprocessing module 10. The main function of this module is to organize the filing data when the user installs photovoltaic power generation. First, collect the filing data submitted by the user when installing the photovoltaic power generation system, and extract 5 types of data including installed capacity (3kw), energy storage capacity (0.45kw), deployment location and photovoltaic panel installation inclination (due east, inclination 15 degrees), and photovoltaic panel area (18 square meters); then, according to the installed capacity and photovoltaic type and brand, convert it into the actual area of ​​the photovoltaic panel and record it; then, according to the relevant national standards and industry standards, analyze the relationship between the data: it meets the ratio relationship between installed capacity and energy storage capacity in the national standard; record the five data of installed capacity, energy storage capacity, deployment location, photovoltaic panel installation inclination and photovoltaic panel area.

[0168] Photovoltaic panel area verification module 20 based on satellite remote sensing images. The main function of this module is to analyze the latest high-resolution remote sensing images of the user's area to achieve accurate measurement of the photovoltaic panel area, so as to verify whether the registered data is correct. First, it is necessary to collect multiple high-resolution remote sensing images of the user's area, and the resolution must be greater than 1 meter (this embodiment uses high-resolution satellite data); then, according to the address provided by the user's registered data, convert it into longitude and latitude coordinates, and mark the longitude and latitude coordinates in the high-resolution image; then, according to the photovoltaic panel area data (18 square meters) recorded by module 10, circle the 10 times area around the positioning coordinates in the high-resolution image, that is, observe the image area of ​​180 square meters; then, for the multiple high-resolution remote sensing images collected, the image of the 180 square meter area is intercepted to obtain n images, which are recorded as IM n ; At the same time, the sensor parameter data of the image is extracted, mainly the shooting angle (the position of the satellite); then, according to the image shooting time and the position of the satellite, the position and angle relationship of the sun, photovoltaic panel and satellite are analyzed: the high-resolution remote sensing data of the high-resolution remote sensing satellite within the reflection angle of the sunlight is eliminated; finally, in the remaining remote sensing images, a deep learning model is constructed to extract the visual area of ​​the photovoltaic panel (the area in the above figure), and the actual area of ​​the photovoltaic panel is converted according to the inclination angle of the photovoltaic panel and the satellite shooting angle (the actual area converted in the embodiment is 21.25 square meters). Compared with the registered data of 18 square meters, the error is greater than 5%, and it is marked as "the user registered data is incorrect".

[0169] Module 30 for comparing and analyzing electricity consumption behaviors before and after photovoltaic installation. The main function of this module is to analyze the characteristics of electricity consumption behaviors of users before and after installing photovoltaic panels. First, the electricity consumption behaviors of the user in one year before installing the photovoltaic power generation system are counted, including the time period and electricity consumption; at the same time, the weather information of the time period before the installation of photovoltaics is collected and counted, including temperature, humidity, air pressure, rainfall, and snowfall; then, a random forest model is constructed, and a sample set is made for training with various data of time periods in previous years; in the same way, the weather and electricity consumption characteristics of the time periods after the installation of photovoltaic panels are counted; then, according to the registered photovoltaic system data, the weather data (including sunshine duration) is combined to convert the installation time of photovoltaic panels. After installing the photovoltaic panels, the daily power generation; next, it is known that the user's photovoltaic installed capacity is 3kw. At the beginning of the peak power consumption period of the day (before 18:00 at night), the remaining energy storage capacity is 0.4, and the energy storage capacity is 15% of the photovoltaic installed capacity. It can be seen that 0.4<0.15*3; using the model in step 3.3 (training data before installing photovoltaic panels), predict the power consumption of all dates and all time periods after installing photovoltaic panels; suppose that the user is in a certain period T (18:00-24:00 at night), the predicted power consumption is 5.2kw; the power consumption of the power grid P e The query result is 1.8kw; it can be seen that P h Much larger than P e +0.15P d , marked as "incorrect filing data".

[0170] "Extreme weather" electricity consumption behavior analysis module 40. This module is mainly used to further determine the electricity consumption behavior of users after the energy storage power is used up. First, if a certain day is defined as "extreme weather", the electricity consumption data of the user on that day is analyzed; if the user does not use the grid distribution, it means that the user is still using the energy storage power; then, the extreme weather collected in the embodiment lasts for a total of 8 days, and the continuous electricity consumption characteristics of the user for 8+1 days (including the night before the extreme weather) can be further analyzed; finally, by comparing P m and P j The size is compared to obtain the user registration status identifier. In this embodiment, P m The value is 7.6kw, P j The value is 10.2, P m Much smaller than P j , so it is marked as "incorrect filing data".

[0171] Photovoltaic power generation filing data verification module 50. This module comprehensively analyzes whether the photovoltaic power generation filing data is incorrect based on the marking results of the above modules. First, in this embodiment, modules 20, 30, and 40 are all marked as "filing data is incorrect", and the final definition is "filing data is incorrect". After the staff's on-site verification, the user has the following situation: adding photovoltaic panels without filing and connecting to the network by themselves; the verification result is consistent with the verification method adopted by the present invention.

[0172] Professionals may further appreciate that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to functionality. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not exceed the scope of the present invention.

[0173] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory, a memory, a read-only memory, an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0174] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by technicians in this technical field within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention.

Claims

1. A method for verifying personal photovoltaic power generation registration data, characterized in that: The steps include: Step 1: Preprocessing of photovoltaic power generation filing data: sort out and screen the filing data of users when installing photovoltaic power generation, and record the user's address, installed capacity, energy storage capacity, deployment location, photovoltaic panel installation angle and actual area of ​​photovoltaic panels as preprocessed filing data; Step 2: Verify the area of ​​photovoltaic panels based on satellite remote sensing images: Analyze the latest high-resolution remote sensing images of the user's area to accurately measure the area of ​​photovoltaic panels, so as to verify whether the pre-processed filing data obtained in step 1 is correct; Step 3: Compare and analyze the electricity consumption behavior before and after the photovoltaic installation: Analyze the electricity consumption characteristics of users before and after the installation of photovoltaic panels. If the analysis shows that the electricity consumption increases after the photovoltaic installation, it is judged that "the registered data is correct", otherwise it is judged that "the registered data is incorrect"; Step 4: "Extreme Weather" electricity consumption behavior analysis: Determine the user's electricity consumption behavior after the energy storage power is used up. If the power consumption during photovoltaic power generation is much less than the power consumption during grid power supply, it is marked as "the registered data is incorrect", otherwise it is marked as "the registered data is correct"; Step 5: Photovoltaic power generation registration data verification: Based on the marking results obtained in steps 2, 3 and 4, comprehensively analyze whether the photovoltaic power generation registration data is incorrect.

2. The method for verifying personal photovoltaic power generation registration data according to claim 1, characterized in that: Step 1 specifically includes: Step 1.1: Collect the filing data submitted by the user when installing the photovoltaic power generation system, and extract 5 types of data including user address, installed capacity, energy storage capacity, deployment location and photovoltaic panel installation inclination; Step 1.2: Convert the installed capacity, photovoltaic type and brand into the actual area of ​​photovoltaic panels and record it; Step 1.3: According to the relevant national standards and industry standards, the collected filing data and the converted actual area of ​​PV panels are screened to select the data that meets the standards; Step 1.4: After the final screening is qualified, the six data including user address, installed capacity, energy storage capacity, deployment location, photovoltaic panel installation angle and actual area of ​​photovoltaic panels will be recorded as pre-processed filing data.

3. The method for verifying personal photovoltaic power generation registration data according to claim 1, characterized in that: Step 2 specifically includes: Step 2.1, collect multiple high-resolution remote sensing images of the user's area, with a resolution greater than 1 meter; Step 2.2: Convert the user address provided in the user filing data into longitude and latitude coordinates; Step 2.3, mark the longitude and latitude coordinates in the high-resolution remote sensing image; Step 2.4: Based on the actual area of ​​the photovoltaic panel obtained in step 1, record it as Area, and circle the area 10 times the actual area of ​​the photovoltaic panel around the positioning coordinates in the high-resolution image, that is, observe the image area of ​​10Area; Step 2.5: For the collected high-resolution remote sensing images, intercept the images of 10Area areas to obtain n images, which are recorded as IM n ; Step 2.6, extracting sensor parameter data of the image, wherein the sensor parameter data includes image shooting time and image shooting angle; Step 2.7: Analyze the position and angle relationship of the sun, photovoltaic panel, and satellite according to the image shooting time and image shooting angle: if the satellite is within the reflection angle of sunlight, the image shot at that time point will be eliminated; Step 2.8: In the remaining remote sensing images, build a deep learning model to extract the visual area of ​​the photovoltaic panels; Step 2.9: Convert the visual area into the actual area of ​​the photovoltaic panel according to the photovoltaic panel installation angle and the image shooting angle; Step 2.10, record the maximum value of the actual area of ​​the photovoltaic panel calculated from multiple images, and record it as realArea; Step 2.11, compare the Area and realArea values. If the difference between the two is less than 5%, it is marked as "the registered data is correct".

4. The method for verifying personal photovoltaic power generation registration data according to claim 3, characterized in that: Step 2 also includes: Step 2.12: If the difference between Area and realArea is greater than 5%, it is considered that other registered photovoltaic panels may be included in the statistics, and further processing is required: a) First, count whether there are other photovoltaic power generation records within a certain range of photovoltaic record coordinates. If not, remove other record interferences and define it as "wrong record data"; b) If there are other registered data within a certain range, repeat steps 2.4-2.5 for the other registered data to extract 10 times the area of ​​all registered photovoltaic panel images within the range; c) For the realArea selected in step 2.10, backtrack to find the corresponding high-resolution remote sensing image; d) extracting the photovoltaic panel areas of all registered users in the region based on the remote sensing high-resolution image found in step c); e) Compare the calculated photovoltaic panel areas of all registered users with the sum of the registered data. If the error is less than 5%, the initial user is marked as "registered data is correct"; f) If not, mark it as "the registered data is incorrect".

5. The method for verifying personal photovoltaic power generation registration data according to claim 1, characterized in that: Step three specifically includes: Step 3.1, statistics the user's electricity consumption behavior in the year before the installation of the photovoltaic power generation system, including electricity consumption time period and electricity consumption; Step 3.2: Collect and compile weather information for each time period in the year before the installation of photovoltaic power generation, including temperature, humidity, air pressure, rainfall, and snowfall; Step 3.3: Build a random forest model, and use the various data from the previous year to make a sample set for training; Step 3.4: Count the weather and electricity consumption characteristics in different time periods after the installation of photovoltaic panels; Step 3.5: Calculate the daily power generation after installing the photovoltaic panels based on the registered photovoltaic system data and weather data; Step 3.6: Assume that the installed capacity of photovoltaic power generation of a certain user is P d At the beginning of the peak power consumption period of the day, the remaining energy storage capacity P s , assuming that the energy storage capacity is 15% of the photovoltaic installed capacity, then P s <0.15P d ; Step 3.7: Use the random forest model in step 3.3 to predict the electricity consumption on all dates and in all time periods after the installation of the photovoltaic panels; let the predicted electricity consumption of the user in a certain time period T be P h ; Step 3.6: Assume that the power consumption of the user in the T period is P e , photovoltaic power consumption is P p , then P p <P s ; Step 3.7, then P e +P p is the actual power consumption of the user, P e +P p <P e +0.15P d ; Step 3.8: Compare P h and P e +0.15P d , if P h <P e +0.15P d , it indicates that the registered photovoltaic power generation is redundant, or the power consumption increases after the user installs photovoltaic power generation; this situation is marked as "registered data is correct"; Step 3.9: If P h Much larger than P e +0.15P d , such as P h >1.5(P e +0.15P d ), this situation shows that after the user installed photovoltaic power generation, the electricity consumption dropped sharply, which is marked as "the registered data is wrong", where 1.5 is the empirical coefficient, indicating that the former is 1.5 times greater than the latter, which is used to quantify "far greater than".

6. The method for verifying personal photovoltaic power generation registration data according to claim 1, characterized in that: Step 4 specifically includes: Step 4.1: If a day is defined as "extreme weather", analyze the electricity consumption data of users on that day; Step 4.2: If the user does not use grid power distribution, it means that the user is still using energy storage power; Step 4.3: Assume that the extreme weather lasts for n days in total, and further analyze the continuous electricity consumption characteristics of users on day n+1: a) Assume that in the n+1 days, the first m days use photovoltaic power supply, the kth day uses hybrid power supply, and the next j days use grid power distribution, where k = m+1; b) Assume that the average power consumption in the previous m days is P m , the energy storage power P used on the kth day k1 ; c) The average actual power consumption in the next j days can be monitored and is set as P j ; d) The power distribution consumption of the power grid on the kth day can also be obtained, set as P k2 ; e) Get mP m +P k1 <0.15P d ; f) Assuming that the power consumption on the kth day is close to the average power consumption on the next j days, estimate P k1 Equal to P j -P k2 ; g) Estimate the daily energy storage power consumption in the previous m days: h) Compare P m and P j Size relationship: If Pm>P j , indicating that the power consumption during photovoltaic power generation is large, marked as "the registered data is correct"; i) If P m Much smaller than P j , such as 1.5P m <P j , it means that the data of photovoltaic power generation is far less than the power consumption of grid power supply, and it is marked as "the registered data is wrong".

7. The method for verifying personal photovoltaic power generation registration data according to claim 1, characterized in that: Step 5 specifically includes: Step 5.1: If Step 2, Step 3, and Step 4 are all marked as "Incorrect filing data", the final definition is "Incorrect filing data", and staff are required to conduct on-site verification; Step 5.2: If any two of the steps 2, 3, and 4 are marked as "incorrect data on record", you will need to contact us by phone for verification and confirm it based on the latest high-resolution image visual interpretation; Step 5.3: If any of the steps 2, 3, and 4 is marked as "incorrect registration data", mark the user registration status as "pending" and repeat steps 1 to 4 after a certain period for analysis and verification.

8. A personal photovoltaic power generation registration data verification system, characterized in that: include: Photovoltaic power generation filing data preprocessing module, used to sort out and filter the filing data of users when installing photovoltaic power generation, and record the user's address, installed capacity, energy storage capacity, deployment location, photovoltaic panel installation angle and actual area of ​​photovoltaic panels as preprocessed filing data; The photovoltaic panel area verification module based on satellite remote sensing images is used to analyze the latest high-resolution remote sensing images of the user's area to achieve accurate measurement of the photovoltaic panel area, so as to verify whether the pre-processed filing data obtained in step 1 is correct; The module for comparing and analyzing electricity consumption behavior before and after photovoltaic installation is used to analyze the electricity consumption characteristics of users before and after installing photovoltaic panels. If the analysis shows that the electricity consumption increases after photovoltaic installation, it is judged that "the registered data is correct", otherwise it is judged that "the registered data is incorrect"; The "extreme weather" electricity consumption behavior analysis module is used to determine the user's electricity consumption behavior after the energy storage power is used up. If the power consumption during photovoltaic power generation is much less than the power consumption during grid power supply, it is marked as "the registered data is incorrect", otherwise it is marked as "the registered data is correct"; The photovoltaic power generation registration data verification module is used to comprehensively analyze whether the photovoltaic power generation registration data is incorrect based on the marking results obtained by the photovoltaic panel area verification module based on satellite remote sensing images, the electricity consumption behavior comparison analysis module before and after photovoltaic installation, and the "extreme weather" electricity consumption behavior analysis module.

9. The personal photovoltaic power generation record data verification system according to claim 8, characterized in that: The photovoltaic power consumption behavior comparison and analysis module before and after installation is specifically used for: Step 3.1, statistics the user's electricity consumption behavior in the year before the installation of the photovoltaic power generation system, including electricity consumption time period and electricity consumption; Step 3.2: Collect and compile weather information for each time period in the year before the installation of photovoltaic power generation, including temperature, humidity, air pressure, rainfall, and snowfall; Step 3.3: Build a random forest model, and use the various data from the previous year to make a sample set for training; Step 3.4: Count the weather and electricity consumption characteristics in different time periods after the installation of photovoltaic panels; Step 3.5: Calculate the daily power generation after installing the photovoltaic panels based on the registered photovoltaic system data and weather data; Step 3.6: Assume that the installed capacity of photovoltaic power generation of a certain user is P d At the beginning of the peak power consumption period of the day, the remaining energy storage capacity P s , assuming that the energy storage capacity is 15% of the photovoltaic installed capacity, then P s <0.15P d ; Step 3.7: Use the random forest model in step 3.3 to predict the electricity consumption on all dates and in all time periods after the installation of the photovoltaic panels; let the predicted electricity consumption of the user in a certain time period T be P h ; Step 3.6: Assume that the power consumption of the user in the T period is P e , photovoltaic power consumption is P p , then P p <P s ; Step 3.7, then P e +P p is the actual power consumption of the user, P e +P p <P e +0.15P d ; Step 3.8: Compare P h and P e +0.15P d , if P h <P e +0.15P d , it indicates that the registered photovoltaic power generation is redundant, or the power consumption increases after the user installs photovoltaic power generation; this situation is marked as "registered data is correct"; Step 3.9: If P h Much larger than P e +0.15P d , such as P h >1.5(P e +0.15P d ), this situation shows that after the user installed photovoltaic power generation, the electricity consumption dropped sharply, which is marked as "the registered data is wrong", where 1.5 is the empirical coefficient, indicating that the former is 1.5 times greater than the latter, which is used to quantify "far greater than".

10. The personal photovoltaic power generation record data verification system according to claim 8, characterized in that: The "extreme weather" electricity consumption behavior analysis module is specifically used for: Step 4.1: If a day is defined as "extreme weather", analyze the electricity consumption data of users on that day; Step 4.2: If the user does not use grid power distribution, it means that the user is still using energy storage power; Step 4.3: Assume that the extreme weather lasts for n days in total, and further analyze the continuous electricity consumption characteristics of users on day n+1: b) Assume that in the n+1 days, the first m days use photovoltaic power supply, the kth day uses hybrid power supply, and the next j days use grid power distribution, where k = m+1; b) Assume that the average power consumption in the previous m days is P m , the energy storage power P used on the kth day k1 ; c) The average actual power consumption in the next j days can be monitored and is set as P j ; d) The power distribution consumption of the power grid on the kth day can also be obtained, set as P k2 ; e) Get mP m +P k1 <0.15P d ; f) Assuming that the power consumption on the kth day is close to the average power consumption on the next j days, estimate P k1 Equal to P j -P k2 ; g) Estimate the daily energy storage power consumption in the previous m days: h) Compare P m and P j Size relationship: If Pm>P j , indicating that the power consumption during photovoltaic power generation is large, and marked as "the registered data is correct"; ii) If P m Much smaller than P j , such as 1.5P m <P j , it means that the data of photovoltaic power generation is far less than the power consumption of grid power supply, and it is marked as "the registered data is wrong".

Images