Intelligent monitoring method and system for tidal flat photovoltaic power generation system

Abstract

The present invention relates to the field of photovoltaic system monitoring technology, and specifically to an intelligent monitoring method and system for a tidal flat photovoltaic power generation system, the method comprising: obtaining historical monitoring data of the tidal flat photovoltaic power generation system; drawing a line graph with time as the horizontal coordinate and historical monitoring data as the vertical coordinate according to each historical monitoring data, and calculating the tidal flat environmental impact factor according to the line graph and the historical monitoring data; performing clustering processing on multidimensional data by a clustering algorithm to obtain at least two clusters, the multidimensional data including: tidal flat environmental impact factor, impact data and time point; matching the current multidimensional data to be monitored to the corresponding cluster, and obtaining the power system data range of the cluster corresponding to the successful match; judging whether the current power system data to be monitored belongs to the power system data range, if not, the current power system data to be monitored is abnormal. By adopting the present invention, it is possible to monitor whether the power system data is abnormal.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic system monitoring, and in particular to an intelligent monitoring method and system for a tidal flat photovoltaic power generation system. Background Art

[0002] The tidal flat photovoltaic power generation system is a photovoltaic power generation system built on tidal flats (such as beaches, estuaries, etc.). After fully considering factors such as terrain and sunlight direction, photovoltaic panels are set up and laid out in order to maximize power generation efficiency and system stability. However, the tidal flat area has a special environment, and the power generation process may be affected by various environmental factors (such as tidal influence, salt spray corrosion, sediment deposition, etc.), which may cause the loss of power generation equipment and lead to abnormal power system data. Therefore, in order to ensure the power generation efficiency of the tidal flat photovoltaic power generation system, it is necessary to monitor the power system data in combination with historical power data and tidal flat environmental characteristics. Summary of the invention

[0003] In order to solve the technical problem of power generation data monitoring in a tidal flat photovoltaic power generation system, the purpose of the present invention is to provide an intelligent monitoring method for a tidal flat photovoltaic power generation system. The technical solution adopted is as follows:

[0004] Acquire historical monitoring data of the tidal flat photovoltaic power generation system, wherein the historical monitoring data includes: at least two power system data and at least two impact data;

[0005] According to each of the historical monitoring data, a line graph is drawn with time as the horizontal axis and the historical monitoring data as the vertical axis, and a tidal flat environmental impact factor is calculated according to the line graph and the historical monitoring data;

[0006] Performing clustering processing on multidimensional data by a clustering algorithm to obtain at least two clusters, wherein the multidimensional data includes: the tidal flat environment influencing factor, the influencing data and a time point;

[0007] Matching the current multi-dimensional data to be monitored to the corresponding cluster, and obtaining the power system data range of the cluster corresponding to the successful match;

[0008] It is determined whether the current power system data to be monitored is within the power system data range. If not, the current power system data to be monitored is abnormal.

[0009] Calculating the tidal flat environmental impact factor according to the line graph and the historical monitoring data includes:

[0010] Data in power systems In the line chart, get the Data points and Normalized power system data slope between data points;

[0011] Influencing data In the line chart, get the The data points are related to the The standardized influence data slope between data points is Traversing the remaining influencing data, the standardized influencing data slopes having the same number as the influencing data dimensions are obtained;

[0012] Get impact data The power system data The complete intervention degree is obtained by standardizing the complete intervention degree to obtain a standardized complete intervention degree;

[0013] Multiplying each of the standardized impact data slopes by the standardized complete intervention degree and then adding them together to obtain a sum value;

[0014] Subtract the slope of the standardized power system data from the sum and take the absolute value, and use the inverse of the absolute value as the horizontal coordinate of the tidal flat environment impact factor;

[0015] The power system data Middle The tidal flat environmental impact factor of a data point is the value obtained by substituting the abscissa of the tidal flat environmental impact factor into the natural exponential function.

[0016] The process of obtaining the complete intervention degree includes:

[0017] Obtaining the impact data The power system data The specific intervention degree and other impact data on the impact data About the power system data The interference coefficient of intervention degree;

[0018] Calculating the product of the specific intervention degree and the intervention degree interference coefficient;

[0019] The total interference degree is a value obtained by subtracting the product from the specific interference degree.

[0020] The process of obtaining the interference coefficient includes:

[0021] Taking the absolute value of the change rate sequence of the influencing data, and performing standardization processing on the change rate sequence after taking the absolute value to obtain the change rate sequence of the influencing data;

[0022] Calculate the impact data The power system data Specific intervention and impact data The power system data The absolute value of the difference of the specific intervention degree is obtained, and the sum of the absolute value of the difference and the adjustment coefficient is obtained as the denominator;

[0023] Obtain the change rate sequence of the impact data The series of the rate of change of the affected data The mean square error value is divided by the denominator to obtain an intervention value;

[0024] Repeat the above steps to convert the influence data change rate sequence Traverse the other sequences that affect the data change rate in sequence, and get The intervention values, wherein the is the number of influencing data dimensions;

[0025] The intervention coefficient is The value obtained by adding the intervention values ​​and performing range standardization.

[0026] The process of obtaining the specific intervention degree includes:

[0027] The impact data are interpolated using Perform filling and obtain a normalized value of the filling quantity;

[0028] Calculate the impact data change rate series The power system data change rate series The correlation coefficient between

[0029] The specific intervention degree is obtained by dividing the absolute value of the correlation coefficient by the normalized value.

[0030] The process of acquiring the sequence of the influence data change rate includes:

[0031] Obtain the slope of adjacent data points in a line graph with time as the horizontal axis and the influencing data as the vertical axis, the element value of the influencing data change rate sequence is the slope, and the number of influencing data is In the case of , the number of elements in the sequence affecting the data change rate is .

[0032] The clustering algorithm is a K-Means clustering algorithm, wherein the number of clusters is obtained by the elbow method, and the clustering distance is the Euclidean distance between the multidimensional data.

[0033] Wherein, the method further comprises:

[0034] When the current multi-dimensional data to be monitored is not matched to the corresponding cluster, the current multi-dimensional data to be monitored is directly included in the cluster with the closest spatial distance thereto.

[0035] The power system data includes: current data, voltage data and power generation data; the influencing data includes: light intensity data, temperature data, air humidity data and tidal water level data.

[0036] The present invention also provides an intelligent monitoring system for a tidal flat photovoltaic power generation system, which applies the above method and comprises:

[0037] An acquisition module, used to acquire historical monitoring data of the tidal flat photovoltaic power generation system, wherein the historical monitoring data includes: at least two power system data and at least two impact data;

[0038] A calculation module, used for drawing a line graph with time as the horizontal axis and the historical monitoring data as the vertical axis according to each of the historical monitoring data, and calculating the tidal flat environment impact factor according to the line graph and the historical monitoring data;

[0039] A clustering module, used for performing clustering processing on multidimensional data by a clustering algorithm to obtain at least two clusters, wherein the multidimensional data includes: the tidal flat environment influencing factors, the influencing data and time points;

[0040] A matching module, used for matching the current multi-dimensional data to be monitored to the corresponding cluster, and obtaining the power system data range of the cluster corresponding to the successful match;

[0041] The judgment module is used to judge whether the current power system data to be monitored belongs to the power system data range. If not, the current power system data to be monitored is abnormal.

[0042] The present invention has the following beneficial effects:

[0043] The present invention provides an intelligent monitoring method and system for a tidal flat photovoltaic power generation system, which obtains historical monitoring data of the tidal flat photovoltaic power generation system, wherein the historical monitoring data includes at least two power system data and at least two impact data; a line graph with time as the horizontal coordinate and the historical monitoring data as the vertical coordinate is drawn according to each of the historical monitoring data, and the tidal flat environmental impact factor is calculated according to the line graph and the historical monitoring data; at least two clusters are obtained by clustering multidimensional data through a clustering algorithm, wherein the multidimensional data includes the tidal flat environmental impact factor, the impact data and the time point; the current multidimensional data to be monitored is matched to the corresponding cluster, and the power system data range of the cluster corresponding to the successful match is obtained; and it is determined whether the current power system data to be monitored belongs to the power system data range, and if not, the current power system data to be monitored is abnormal. Compared with the prior art, in the clustering process, the data with positive and negative impacts on the power system data are comprehensively analyzed and evaluated, which further expands the consideration range of the impact data and improves the accuracy of the impact of the historical data clustering results on the power system data. Evaluating current monitoring data through historical data improves the adaptability of the evaluation process to different tidal flat areas and reduces errors in monitoring data evaluation results caused by different natural environmental conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 A flow chart of an intelligent monitoring method for a tidal flat photovoltaic power generation system provided by one embodiment of the present invention;

[0046] Figure 2 A flow chart of a calculation process of a tidal flat environmental impact factor provided by an embodiment of the present invention;

[0047] Figure 3 A flowchart of a process for calculating the total intervention degree provided by one embodiment of the present invention;

[0048] Figure 4 A flowchart of a calculation process of an intervention degree interference coefficient provided by an embodiment of the present invention;

[0049] Figure 5 A flowchart of a process for calculating a specific intervention degree provided by one embodiment of the present invention;

[0050] Figure 6 A schematic diagram of an intelligent monitoring system for a tidal flat photovoltaic power generation system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the intelligent monitoring method and system of a tidal flat photovoltaic power generation system proposed by the present invention, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0052] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0053] The specific scheme of the intelligent monitoring method for a tidal flat photovoltaic power generation system provided by the present invention is described in detail below with reference to the accompanying drawings.

[0054] See also Figure 1 , which shows a flow chart of an intelligent monitoring method for a tidal flat photovoltaic power generation system provided by an embodiment of the present invention, the method comprising:

[0055] S101. Acquire historical monitoring data of the tidal flat photovoltaic power generation system, the historical monitoring data comprising: at least two power system data and at least two impact data;

[0056] The environmental characteristics of different tidal flats have different impacts on photovoltaic power generation systems. In order to analyze the unique environmental characteristics of the current tidal flat photovoltaic power generation system, it is necessary to analyze it in combination with the historical data of the current power generation system.

[0057] By installing sensors such as current sensors, voltage sensors, temperature sensors, etc. on key equipment such as photovoltaic modules, inverters, junction boxes, etc., key historical data of the tidal flat photovoltaic power generation system, such as voltage, current, power generation, power generation, temperature, light intensity, etc., can be obtained. Through weather stations or some self-installed water level sensors, humidity sensors, etc., meteorological information such as historical high tide levels and air humidity can be obtained.

[0058] For the convenience of subsequent explanation, the parameters that directly reflect the power generation system information are referred to as power system data, and the data that directly or indirectly affects the power system data is referred to as impact data. The power system data includes: current data, voltage data, and power generation data, and the impact data includes: light intensity data, temperature data, air humidity data, and tidal water level data.

[0059] These power system data and impact data can be transmitted to the data processing center through wireless communication (such as LoRa, NB-IoT, 5G, etc.), and the power system data or impact data with less data volume can be filled by interpolation method, so that the power system data and the impact data can correspond to each other at each time point. The interpolation method includes but is not limited to Lagrange interpolation method and least squares interpolation method.

[0060] The number of dimensions of power system data and impact data is at least two. For each dimension of data, there can be The number of dimensions may refer to the type of the power system data or the impact data. For example, when the power system data only includes voltage data and current data, the number of dimensions of the power system data is 2.

[0061] S102. Draw a line graph with time as the horizontal axis and the historical monitoring data as the vertical axis according to each of the historical monitoring data, and calculate the tidal flat environmental impact factor according to the line graph and the historical monitoring data;

[0062] A line graph with time as the horizontal axis and the power system data as the vertical axis can be called a power system data line graph. For example, a line graph with time as the horizontal axis and voltage data as the vertical axis can be called a voltage data line graph.

[0063] A line graph with time as the horizontal axis and the impact data as the vertical axis may be called an impact data line graph. For example, a line graph with time as the horizontal axis and the light intensity data as the vertical axis may be called a light intensity data line graph.

[0064] The calculation of the tidal flat environmental impact factor based on the line graph and the historical monitoring data can be referred to the specific embodiments below, which will not be described in detail here.

[0065] S103. Performing clustering processing on the multidimensional data by a clustering algorithm to obtain at least two clusters, the multidimensional data comprising: the tidal flat environment influencing factors, the influencing data and time points;

[0066] In the multidimensional data space, the abnormality of the data is reflected by the above three types of data: in a certain tidal flat, the range of changes affecting the data is usually determined and has certain rules, that is, the tidal flat environmental impact factors of each power system data point at different times are combined with weather factors, and the power system data obtained has a corresponding data range. It should be noted here that the historical monitoring data are all normal, so the multidimensional data in step S103 are also normal, and the power system data corresponding to the cluster are also normal. Only in this way can it be used as a basis for subsequent judgment of whether the current power system data to be monitored is abnormal.

[0067] Wherein, when clustering the multidimensional data by a clustering algorithm, the tidal flat environmental impact factor, the impact data and the time point constitute a multidimensional data, and the multidimensional data can be understood as a vector. The clustering algorithm is a K-Means clustering algorithm, wherein the number of clusters is obtained by an elbow method, and the cluster distance is the Euclidean distance between the multidimensional data.

[0068] S104. Match the current multi-dimensional data to be monitored to the corresponding cluster, and obtain the power system data range of the cluster corresponding to the successful match;

[0069] Wherein, the method further comprises:

[0070] When the current multi-dimensional data to be monitored is not matched to the corresponding cluster, the current multi-dimensional data to be monitored is directly included in the cluster with the closest spatial distance thereto.

[0071] It can be seen that the K-Means clustering algorithm is an unsupervised learning clustering algorithm based on the center. It groups samples into several clusters through iteration so that the sum of the distances between each sample and the center or mean of the class to which it belongs is minimized. The process of matching the multidimensional data to be monitored to the corresponding cluster is to first find which cluster has the shortest distance from the center of the multidimensional data to be monitored. The cluster corresponding to the shortest distance is successfully matched with the multidimensional data to be monitored.

[0072] S105. Determine whether the current power system data to be monitored is within the power system data range; if not, the current power system data to be monitored is abnormal.

[0073] When the power system data of any dimension in the current power system data to be monitored does not belong to the power system data range, it is considered that the current power system data to be monitored is abnormal.

[0074] Figure 2 A flowchart of a calculation process of a tidal flat environmental impact factor provided by an embodiment of the present invention, wherein the calculation process of the tidal flat environmental impact factor according to the line graph and the historical monitoring data includes:

[0075] S201. Data in power system In the line chart, get the Data points and Normalized power system data slope between data points;

[0076] Power system data The line graph is a graph with time as the horizontal axis and the power system data as the horizontal axis. A line chart with the data values ​​as the vertical axis.

[0077] Data in power systems In the line chart, Data points and After the slopes between the data points are standardized, the standardized power system data slope is obtained. Slope standardization is to standardize the slope using the range standardization method.

[0078] S202. Influencing data In the line chart, get the The data points are related to the The standardized influence data slope between data points is Traversing the remaining influencing data, the standardized influencing data slopes having the same number as the influencing data dimensions are obtained;

[0079] This step means that the number of dimensions that affect the data is If you The standardized impact data slope is obtained. Each dimension data in the impact data obtains the corresponding standardized impact data slope. The standardized impact data slope is obtained in the same manner as the standardized power system data slope.

[0080] S203. Obtaining impact data The power system data A complete intervention degree, and standardizing the complete intervention degree to obtain a standardized complete intervention degree;

[0081] The calculation process of the complete intervention degree will be described in detail in the subsequent embodiments and will not be repeated here.

[0082] S204. multiplying each of the standardized impact data slopes by the standardized complete intervention degree and then adding them together to obtain a sum;

[0083] The above step S204 is expressed by the formula:

[0084]

[0085] in, is the standardized complete intervention degree, Influencing data In the line chart, get the The data points are related to the Normalization between data points affects the slope of the data.

[0086] S205. Subtract the slope of the standardized power system data from the sum value and take the absolute value, and use the inverse of the absolute value as the horizontal coordinate of the tidal flat environmental impact factor;

[0087] The above step S205 is expressed by the formula:

[0088]

[0089] in, Indicates the power system data In the line chart, Data points and After the slopes between the data points are standardized, the standardized power system data slope is obtained, and the sum value is obtained in the above step S204.

[0090] S206. The power system data Middle The tidal flat environmental impact factor of a data point is the value obtained by substituting the abscissa of the tidal flat environmental impact factor into the natural exponential function.

[0091] The power system data Middle The tidal flat environmental impact factor of each data point can be expressed as follows:

[0092]

[0093] in, () is the natural exponential function, () The horizontal coordinates of the factors affecting the tidal flat environment are in brackets.

[0094] The slope calculation process of the influencing data is adjusted by using the complete intervention degree of each influencing data as a coefficient. In the line chart, the size of the slope reflects the size of the positive growth rate of the data. Shows power system data The positive growth rate of After the difference is made, the difference between the positive growth rate of the impact data after the full intervention degree adjustment and the current power system data growth rate is obtained. The smaller the value, the higher the power system data at the current time point. The more obvious the intervention of the affected data is, the more logical the relationship is adjusted by using the exp() function and the value range is set to the interval [0,1]. Represents power system data Middle The tidal flat environmental impact factors of each data point.

[0095] To analyze the interference of other influencing data and current influencing data on a specific power system data, the difference in data change rate and specific intervention degree can be analyzed together: the smaller the difference in the change rate of the influencing data in each time series interval, the higher the correlation between the influencing data; under the current conditions, if the specific intervention degrees of two influencing data on the power system data are closer, it means that the two types of influencing data have similar intervention processes on the current power system data; therefore, the specific intervention degree of the influencing data on the power system data is subject to the interference of another influencing data. Figure 3 A flowchart of a process for calculating a complete intervention degree provided by an embodiment of the present invention, wherein the process for obtaining the complete intervention degree includes:

[0096] S301. Obtain the impact data The power system data The specific intervention degree and other impact data on the impact data About the power system data The interference coefficient of intervention degree;

[0097] The process of obtaining the specific intervention degree and the intervention degree interference coefficient will be described in detail in subsequent embodiments and will not be repeated here.

[0098] S302. Calculate the product of the specific intervention degree and the intervention degree interference coefficient;

[0099] S303. The total interference degree is the value of the specific interference degree minus the product.

[0100] Specifically, the process of obtaining the complete interference degree is:

[0101]

[0102] in, Indicates the impact data The power system data The full degree of intervention, Indicates other data that affects the data in question About the power system data The interference coefficient of the intervention degree, Indicates the impact data The power system data specific level of intervention.

[0103] Among them, the impact data The power system data The greater the specific intervention, and other impact data (except for the impact data (other than the impact data) About the power system data The smaller the interference coefficient, the greater the impact data The power system data Full intervention The bigger.

[0104] Figure 4 A flowchart of a calculation process of an intervention degree interference coefficient provided by an embodiment of the present invention, wherein the calculation process of the intervention degree interference coefficient includes:

[0105] S401. Taking the absolute value of the change rate sequence of the influencing data, and performing standardization processing on the change rate sequence after taking the absolute value to obtain the change rate sequence of the influencing data;

[0106] The process of acquiring the sequence of the influence data change rate includes:

[0107] Obtain the slope of adjacent data points in a line graph with time as the horizontal axis and the influencing data as the vertical axis, the element value of the influencing data change rate sequence is the slope, and the number of influencing data is In the case of , the number of elements in the sequence affecting the data change rate is .

[0108] The impact data The absolute value of the change rate sequence is taken, and the change rate sequence after taking the absolute value is normalized to obtain the change rate sequence of the influencing data .

[0109] The impact data The absolute value of the change rate sequence is taken, and the change rate sequence after taking the absolute value is normalized to obtain the change rate sequence of the influencing data .

[0110] The standardization process refers to processing the data in the change rate sequence using the range standardization method.

[0111] The purpose of taking the absolute value is that for influencing data, such as temperature data, light intensity data, air humidity data, etc., the positive and negative correlation with the power system data may be different, and the positive and negative correlation is an interference factor in the current intervention degree analysis process. Here we only care about the strength of the correlation.

[0112] S402. Calculate the impact data The power system data Specific intervention and impact data The power system data The absolute value of the difference of the specific intervention degree is obtained, and the sum of the absolute value of the difference and the adjustment coefficient is obtained as the denominator;

[0113] The denominator can be expressed as:

[0114]

[0115] in, represents the adjustment coefficient. In this embodiment, =1, Represents the impact data The power system data Specific intervention and impact data The power system data The absolute value of the difference in the specific intervention degree.

[0116] S403. Obtaining the influence data change rate sequence The series of the rate of change of the affected data The mean square error value is divided by the denominator to obtain an intervention value;

[0117] The mean square error value can be used To indicate that, Represents the mean square error calculation function.

[0118] The intervention value can be expressed as:

[0119]

[0120] S404. Repeat the above steps to sequence the influence data change rate. Replace them with other sequences that affect the data change rate, and we get The intervention values, wherein the is the number of influencing data dimensions;

[0121] Here, that is to say, the sequence of the influence data change rate Traverse in sequence The rest of the influence data change rate sequence, to get The intervention value.

[0122] S405. The intervention coefficient is The value obtained by adding the intervention values ​​and performing range standardization.

[0123]

[0124] Indicates other influencing data on the influencing data About Power System Data The interference coefficient of the intervention degree, Represents the range normalization function.

[0125] To analyze the interference of other influencing data and current influencing data on a specific power system data, the difference in data change rate and specific intervention degree can be analyzed together: the smaller the difference in the change rate of the influencing data in each time series interval, the higher the correlation between the influencing data; under the current conditions, if the specific intervention degrees of two influencing data on the power system data are closer, it means that the two types of influencing data have similar intervention processes on the current power system data; therefore, the specific intervention degree of the influencing data on the power system data is subject to the interference of another influencing data.

[0126] In the tidal flat photovoltaic power generation system, the power system data is the data information that can directly reflect the actual power generation situation of the power generation system, and the impact data is the data information that can interfere with the power generation efficiency of the photovoltaic power generation system; to evaluate the operating status of the photovoltaic power generation system, it is necessary to analyze the power generation level of the current power system data under the influence of the current impact data. Therefore, it is necessary to first analyze the impact of the impact data of different dimensions on the power system data. Figure 5 A flowchart of a specific intervention degree calculation process provided by an embodiment of the present invention, wherein the specific intervention degree calculation process includes:

[0127] S501. Using interpolation method to the impact data Perform filling and obtain a normalized value of the filling quantity;

[0128] The interpolation method includes but is not limited to Lagrange interpolation method and least squares interpolation method.

[0129] The normalization method may be a range normalization method, and the range normalization formula is:

[0130]

[0131] Wherein, X is the impact data Fill to get the filling quantity, X min is the minimum value of the filling quantity, X max is the maximum fill quantity, ‌X norm is the normalized value. That is to say, there are as many filling numbers after interpolation as there are dimensions affecting the data. These filling numbers can form a set, X min and X max are the minimum and maximum values ​​of this set respectively.

[0132] Influence data Normalized value for the amount of padding during interpolation.

[0133] S502. Calculate the sequence of the rate of change of the impact data The power system data change rate series The correlation coefficient between

[0134] Affecting data change rate sequence The power system data change rate series The correlation coefficient between To indicate that, Represents the correlation coefficient calculation function.

[0135] The correlation coefficient calculation function may be a Pearson correlation coefficient function.

[0136] S503. Divide the absolute value of the correlation coefficient by the normalized value to obtain the specific intervention degree.

[0137]

[0138] in, Influence data Power system data specific level of intervention.

[0139] For impact data For example, when the power system data changes accordingly during its change process, and the degree of change is closer to the difference between the current influencing data (or the more symmetrical the change difference is), it means that when other influencing data are constant, the current influencing data has a higher degree of intervention in the power system data.

[0140] pass The absolute value of the correlation coefficient between the impact data change rate sequence and the power system data change rate sequence is calculated. The larger the value, the closer the change relationship between the impact data and the power system data is, that is, the higher the degree of intervention of the current impact data on the power system data.

[0141] In addition, since the collection frequency of impact data is usually lower than that of power system data, in order to achieve the mutual comparison process, the data is processed by interpolation. However, the data obtained by interpolation is obtained by simulation operation, and its credibility is relatively low. Therefore, the normalized value of the number of interpolated data of the impact data is used as the denominator. Influence data Power system data specific level of intervention.

[0142] In the process of analyzing the specific intervention degree, the joint effect of other influencing factors on the power system data is not taken into account. That is, the specific intervention degree simulates an impact characteristic that is approximately ideal, which cannot represent the impact of actual influencing data of different dimensions on the power system data. It is also necessary to analyze its complete intervention degree in combination with the change relationship between the influencing data.

[0143] Figure 6 A schematic diagram of an intelligent monitoring system for a tidal flat photovoltaic power generation system provided by an embodiment of the present invention, wherein the intelligent monitoring system applies the above method and includes:

[0144] An acquisition module 601 is used to acquire historical monitoring data of the tidal flat photovoltaic power generation system, wherein the historical monitoring data includes: at least two power system data and at least two impact data;

[0145] A calculation module 602 is used to draw a line graph with time as the horizontal axis and the historical monitoring data as the vertical axis according to each of the historical monitoring data, and calculate the tidal flat environment impact factor according to the line graph and the historical monitoring data;

[0146] A clustering module 603 is used to perform clustering processing on the multidimensional data by a clustering algorithm to obtain at least two clusters, wherein the multidimensional data includes: the tidal flat environment impact factor, the impact data and the time point;

[0147] A matching module 604 is used to match the current multi-dimensional data to be monitored to the corresponding cluster, and obtain the power system data range of the cluster corresponding to the successful match;

[0148] The judgment module 605 is used to judge whether the current power system data to be monitored belongs to the power system data range. If not, the current power system data to be monitored is abnormal.

[0149] The technical features and technical effects of an intelligent monitoring system for a tidal flat photovoltaic power generation system proposed in an embodiment of the present invention are the same as those of the method proposed in an embodiment of the present invention, and will not be described in detail here.

[0150] The present invention has the following beneficial effects:

[0151] The present invention provides an intelligent monitoring method and system for a tidal flat photovoltaic power generation system, which obtains historical monitoring data of the tidal flat photovoltaic power generation system, wherein the historical monitoring data includes at least two power system data and at least two impact data; a line graph with time as the horizontal coordinate and the historical monitoring data as the vertical coordinate is drawn according to each of the historical monitoring data, and the tidal flat environmental impact factor is calculated according to the line graph and the historical monitoring data; at least two clusters are obtained by clustering multidimensional data through a clustering algorithm, wherein the multidimensional data includes the tidal flat environmental impact factor, the impact data and the time point; the current multidimensional data to be monitored is matched to the corresponding cluster, and the power system data range of the cluster corresponding to the successful match is obtained; and it is determined whether the current power system data to be monitored belongs to the power system data range, and if not, the current power system data to be monitored is abnormal. Compared with the prior art, in the clustering process, the data with positive and negative impacts on the power system data are comprehensively analyzed and evaluated, which further expands the consideration range of the impact data and improves the accuracy of the impact of the historical data clustering results on the power system data. Evaluating current monitoring data through historical data improves the adaptability of the evaluation process to different tidal flat areas and reduces errors in monitoring data evaluation results caused by different natural environmental conditions.

[0152] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0153] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. An intelligent monitoring method for a tidal flat photovoltaic power generation system, characterized in that: The method comprises: Acquire historical monitoring data of the tidal flat photovoltaic power generation system, wherein the historical monitoring data includes: at least two power system data and at least two impact data; According to each of the historical monitoring data, a line graph is drawn with time as the horizontal axis and the historical monitoring data as the vertical axis, and a tidal flat environmental impact factor is calculated according to the line graph and the historical monitoring data; Performing clustering processing on multidimensional data by a clustering algorithm to obtain at least two clusters, wherein the multidimensional data includes: the tidal flat environment influencing factor, the influencing data and a time point; Matching the current multi-dimensional data to be monitored to the corresponding cluster, and obtaining the power system data range of the cluster corresponding to the successful match; Determine whether the current power system data to be monitored is within the power system data range, if not, the current power system data to be monitored is abnormal; Calculating the tidal flat environmental impact factor according to the line graph and the historical monitoring data includes: Data in power systems In the line chart, get the Data points and Normalized power system data slope between data points; Influencing data In the line chart, get the The data points are related to the The standardized influence data slope between data points is Traversing the remaining influencing data, the standardized influencing data slopes having the same number as the influencing data dimensions are obtained; Get impact data The power system data The complete intervention degree is obtained by standardizing the complete intervention degree to obtain a standardized complete intervention degree; Multiplying each of the standardized impact data slopes by the standardized complete intervention degree and then adding them together to obtain a sum value; Subtract the slope of the standardized power system data from the sum and take the absolute value, and use the inverse of the absolute value as the horizontal coordinate of the tidal flat environment impact factor; The power system data Middle The tidal flat environmental impact factor of a data point is the value obtained by substituting the abscissa of the tidal flat environmental impact factor into the natural exponential function.

2. The intelligent monitoring method for a tidal flat photovoltaic power generation system according to claim 1, characterized in that: The process of obtaining the full intervention degree includes: Get the impact data The power system data The specific intervention degree and other impact data on the impact data About the power system data The interference coefficient of intervention degree; Calculating the product of the specific intervention degree and the intervention degree interference coefficient; The total interference degree is a value obtained by subtracting the product from the specific interference degree.

3. The intelligent monitoring method for a tidal flat photovoltaic power generation system according to claim 2, characterized in that: The process of obtaining the interference coefficient of the intervention degree includes: (1) taking the absolute value of the change rate sequence of the influencing data, and performing standardization processing on the change rate sequence after taking the absolute value to obtain the change rate sequence of the influencing data; (2) Calculate the impact data The power system data Specific intervention and impact data The power system data The absolute value of the difference of the specific intervention degree is obtained, and the sum of the absolute value of the difference and the adjustment coefficient is obtained as the denominator; (3) Obtaining the sequence that affects the rate of change of data The series of the rate of change of the affected data The mean square error value is divided by the denominator to obtain an intervention value; Repeat the above steps (2) and (3) to convert the influence data change rate sequence Traverse the other sequences that affect the data change rate in sequence, and get The intervention values, wherein the is the number of influencing data dimensions; The intervention coefficient is The value obtained by adding the intervention values ​​and performing range standardization.

4. The intelligent monitoring method for a tidal flat photovoltaic power generation system according to claim 3, characterized in that: The process of obtaining the specific intervention level includes: The impact data are interpolated using Perform filling and obtain a normalized value of the filling quantity; Calculate the impact data change rate series The power system data change rate series The correlation coefficient between The specific intervention degree is obtained by dividing the absolute value of the correlation coefficient by the normalized value.

5. The intelligent monitoring method for a tidal flat photovoltaic power generation system according to claim 4, characterized in that: The acquisition process of the influence data change rate sequence includes: Obtain the slope of adjacent data points in a line graph with time as the horizontal axis and the influencing data as the vertical axis, the element value of the influencing data change rate sequence is the slope, and the number of influencing data is In the case of , the number of elements in the sequence affecting the data change rate is .

6. The intelligent monitoring method for a tidal flat photovoltaic power generation system according to claim 1, characterized in that: The clustering algorithm is a K-Means clustering algorithm, wherein the number of clusters is obtained by the elbow method, and the clustering distance is the Euclidean distance between the multidimensional data.

7. The intelligent monitoring method for a tidal flat photovoltaic power generation system according to claim 1, characterized in that: The method further comprises: When the current multi-dimensional data to be monitored is not matched to the corresponding cluster, the current multi-dimensional data to be monitored is directly included in the cluster with the closest spatial distance thereto.

8. The intelligent monitoring method for a tidal flat photovoltaic power generation system according to claim 1, characterized in that: The power system data includes: current data, voltage data and power generation data, and the impact data includes: light intensity data, temperature data, air humidity data and tidal water level data.

9. An intelligent monitoring system for a tidal flat photovoltaic power generation system, characterized in that: The method according to any one of claims 1 to 8, comprising: An acquisition module, used to acquire historical monitoring data of the tidal flat photovoltaic power generation system, wherein the historical monitoring data includes: at least two power system data and at least two impact data; A calculation module, used for drawing a line graph with time as the horizontal axis and the historical monitoring data as the vertical axis according to each of the historical monitoring data, and calculating the tidal flat environment impact factor according to the line graph and the historical monitoring data; A clustering module, used for performing clustering processing on multidimensional data by a clustering algorithm to obtain at least two clusters, wherein the multidimensional data includes: the tidal flat environment influencing factors, the influencing data and time points; A matching module, used for matching the current multi-dimensional data to be monitored to the corresponding cluster, and obtaining the power system data range of the cluster corresponding to the successful match; A judgment module, used for judging whether the current power system data to be monitored belongs to the power system data range, if not, the current power system data to be monitored is abnormal; Calculating the tidal flat environmental impact factor according to the line graph and the historical monitoring data includes: Data in power systems In the line chart, get the Data points and Normalized power system data slope between data points; Influencing data In the line chart, get the The data points are related to the The standardized influence data slope between data points is Traversing the remaining influencing data, the standardized influencing data slopes having the same number as the influencing data dimensions are obtained; Get impact data The power system data The complete intervention degree is obtained by standardizing the complete intervention degree to obtain a standardized complete intervention degree; Multiplying each of the standardized impact data slopes by the standardized complete intervention degree and then adding them together to obtain a sum value; Subtract the slope of the standardized power system data from the sum and take the absolute value, and use the inverse of the absolute value as the horizontal coordinate of the tidal flat environment impact factor; The power system data Middle The tidal flat environmental impact factor of a data point is the value obtained by substituting the abscissa of the tidal flat environmental impact factor into the natural exponential function.

Images