Intelligent wiring control method for junction box

Abstract

The present invention relates to the field of data processing, and more specifically, to a method for intelligent wiring control of a junction box, the method comprising: obtaining the operating condition data of the sensor of the historical intelligent junction box, clustering based on the historical operating condition data, obtaining the historical data similar to the current operating condition data according to the clustering result, and calculating the temperature change rate under the same operating condition as the historical data; using the time series data to predict the time series sequence of the temperature change rate in the current operating condition data, and calculating the temperature stability deviation at the same time under the same operating condition; calculating the temperature anomaly value according to the historical data and the change rate and the temperature stability deviation similar to the current operating condition data, and judging whether the junction box has a risk according to the change of the continuous temperature anomaly value under the same historical operating condition. The present invention reduces the misjudgment of the single temperature threshold control and improves the accuracy of detection by identifying the deviation between the actual temperature change and the expected change.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and more specifically, to an intelligent wiring control method for a wiring box. Background Art

[0002] Junction box, also known as terminal box or distribution box, is an electrical device used for circuit connection and distribution. It is used in the fields of power, electronics, communications, automation control, etc. to connect cables, wires or terminals of different lines together to achieve circuit connection, distribution, protection and control.

[0003] The existing Chinese patent application document with publication number CN107834977A discloses a photovoltaic module intelligent junction box, including a main circuit, a voltage detection selection unit, a voltage detection unit, a control unit and a power supply unit. The main circuit includes an output unit and multiple input units, each input unit includes a first switch connected in series with the corresponding battery string, and a second switch connected in parallel with the battery string and the first switch; the control unit is connected to the main circuit and is used to control the on and off of each switch. The application document also discloses a control method, including a detection method for judging short-circuit damage of the second switch, which is: S11, control the second switch to turn off; S12, detect the voltage Va at both ends of the second switch and the voltage Vcell of the battery string; S13, when Va≤first preset value, and Vcell>n*Voc, n is a preset value, where Voc is the open-circuit voltage of the photovoltaic module, it is judged that the second switch is in a short-circuit damage state; S14, control the corresponding first switch to turn off. The junction box and control method in the application document have the advantages of safety and reliability.

[0004] In this application document, automatic control of the smart switch is achieved only through voltage detection and logical judgment to ensure the safe and stable operation of the photovoltaic system. At present, the smart junction box collects timing data through temperature sensors, voltage sensors, and current sensors to implement switch control. However, due to the different characteristics of temperature changes under different working conditions, if only relying on setting a fixed safety temperature value to prevent the risk of spontaneous combustion, it may lead to misjudgment or missed judgment. This single temperature threshold control strategy may not be able to accurately respond to all working conditions, thereby increasing the risk of causing spontaneous combustion. Summary of the invention

[0005] In order to solve the problem that due to the different characteristics of temperature changes under different working conditions, a single temperature threshold control strategy may not be able to accurately cope with all working conditions, thereby increasing the risk of causing spontaneous combustion, the present invention provides solutions in the following aspects.

[0006] A method for controlling intelligent wiring of a junction box, comprising: obtaining historical operating condition data of a sensor of an intelligent junction box, wherein the operating condition data includes voltage data, current data, temperature data and time series data; clustering based on the historical operating condition data, obtaining historical data similar to the current operating condition data according to the clustering result, and calculating the current temperature value and the change rate of the temperature under the same operating condition in the historical data; using a time series data prediction method, predicting the time series of the temperature change rate in the current operating condition data, obtaining multiple time periods of the predicted temperature change rate and the temperature change rate under the same operating condition in history, and calculating the temperature stability deviation at the same time under the same operating condition; calculating the temperature anomaly value according to the historical data and the change rate and the temperature stability deviation similar to the current operating condition data, and judging whether the junction box has risks according to the changes of continuous temperature anomaly values ​​under the same operating condition in history;

[0007] The temperature abnormal value Satisfies the following relationship: , where Indicates historical data that is similar to the current operating data. Indicates the conversion rate between the temperature data of the current working condition data and the temperature data of the same working condition in the historical data. Indicates the temperature stability deviation between the predicted stable temperature value of the temperature value of the current working condition data and the stable temperature under the same working condition in history. Represented by natural numbers An exponential function with base .

[0008] The effect is: by obtaining historical data similar to the current operating conditions through cluster analysis and calculating the temperature change rate, the trend and characteristics of temperature changes can be captured more accurately, thereby improving the accuracy of temperature control; using the time series data prediction method, the predicted temperature change rate and multiple time periods of the temperature change rate under the same historical operating conditions are obtained, thereby reducing the misjudgment of the current operating condition data, and combining the temperature anomalies under the same historical operating conditions to improve the risk prediction ability.

[0009] Preferably, obtaining historical data similar to the current operating condition data includes:

[0010] Based on the time series data, the voltage data and the current data are clustered for the first time to obtain multiple clusters, the mean power of the clusters is calculated, the clusters are sorted according to the mean power, the sorting number of each category is obtained, and normalization is performed;

[0011] The time series of sensor data is replaced with a normalized sequence of numbers, and the time points corresponding to the numbers in the sequence of numbers are clustered for the second time to obtain multiple sequence time clusters;

[0012] Calculate the mean of the sequence number in the sequence time clustering cluster, and perform clustering for the third time to obtain multiple sequence mean clustering clusters. Get the corresponding sensor data according to the sequence number in the sequence mean clustering cluster. Calculate the approximation between the sequence number corresponding to the sequence mean clustering cluster where the current collected data is located and the sequence number data corresponding to each sequence mean clustering cluster, and obtain the maximum value of the approximation as the historical data that is similar to the current operating condition data.

[0013] The effect is as follows: by classifying similar operating condition data into one category, the actual operating condition characteristics can be more accurately reflected; by calculating the approximation between the serial number corresponding to the serial number mean clustering cluster where the current collected data is located and the serial number data corresponding to each serial number mean clustering cluster, and selecting the maximum value of the approximation as the historical data similar to the current operating condition data, the misjudgment of the current operating condition data can be effectively reduced; combined with the temperature anomaly analysis under the same historical operating conditions, this method can more accurately identify the real abnormal situation, thereby improving the accuracy of the detection of abnormal temperature of the junction box.

[0014] Preferably, obtaining historical data similar to the current operating condition data also includes:

[0015] The historical data and current operating condition data are converted into feature vectors respectively, the nearest neighbor search algorithm is used to search for the most similar data points, and the cosine similarity between the feature vector of the historical data and the feature vector of the current operating condition data is calculated. The searched data points of the historical data are sorted according to the cosine similarity, and the top N data points with the smallest cosine similarity are selected as the historical data that are similar to the current operating condition data.

[0016] The effect is that through the nearest neighbor search algorithm, the data points most similar to the current working conditions can be quickly found in the historical data, which can reduce the time of searching for similar items in complex data sets, optimize the decision-making process, and improve the response speed.

[0017] Preferably, the conversion rate includes:

[0018] Align the current working condition data with the historical working condition data on the time axis, determine the starting time, calculate the time interval between the current time and the starting time of the current working condition data, find the corresponding time point in the aligned historical working condition data according to the time interval, and obtain the average temperature value at the same time point as the expected temperature;

[0019] Get the actual temperature at the time point corresponding to the current operating condition data, calculate the difference between the actual temperature and the expected temperature, and get the temperature change rate.

[0020] The effect is that by aligning the current operating condition data with the historical operating condition data on the time axis, the temperature data under the same or similar conditions is ensured, thereby reducing the error caused by time dislocation and improving the accuracy of temperature prediction.

[0021] Preferably, the change rate further includes:

[0022] The temperature data in the current working condition data and the temperature data of multiple periods under the same working condition in history are obtained, and the temperature data of multiple periods under the same working condition in history are fitted using the least square method to obtain a fitting curve, and the average value of the same time point in the fitting curve of multiple time periods is calculated as the expected stable temperature under the same working condition in history;

[0023] Obtain the actual temperature at the same time point of the current working condition data and the expected stable temperature, calculate the difference between the actual temperature and the expected temperature, and obtain the temperature change rate.

[0024] Preferably, the temperature stability deviation includes:

[0025] The average value of the temperature change rate at the same time point under the same working conditions in multiple periods of history is taken as the stable temperature, and the difference between the stable temperature at the same time and the temperature change rate of the current working condition data is taken as the temperature stability deviation.

[0026] The effect is that by evaluating the temperature change trend under the same historical operating conditions and comparing it with the temperature change rate of the current operating condition data, it can effectively identify whether the temperature fluctuation is within the normal range, thereby enhancing the real-time monitoring capability of the equipment operating status.

[0027] Preferably, the temperature stability deviation further includes:

[0028] Align the current working condition data with the historical data on the time axis, obtain the temperature data of the historical data and the temperature data of the current working condition at the same time point, and preset a sliding window, respectively calculate the first-order difference values ​​of the temperature data in the sliding window corresponding to the current working condition data and the historical data, and obtain the first-order difference sequence of the temperature data in the window corresponding to the current working condition data and the historical data, and respectively use the mean of the first-order difference sequence of the preset sliding window as the stability of the central data point in the preset window, and when the stability is less than the stability threshold, it is used as low stable temperature data, otherwise, it is high stable temperature data, and respectively obtain the time periods corresponding to the continuous low stable temperature data in the current working condition data and the historical data, wherein there is at least one time period corresponding to the low stable temperature data;

[0029] Calculate the ratio of the temperature difference at the beginning and end of the preset sliding window of the time period corresponding to the low stable temperature data in the current working condition data and the historical data to the time interval respectively, and obtain the temperature change rate of the current working condition data and the historical data respectively. Calculate the difference between the average of the temperature change rates of multiple low stable temperature data corresponding to the time period of the current working condition data and the average of the temperature change rates of multiple low stable temperature data corresponding to the time period of the historical data respectively, and obtain the temperature stability deviation.

[0030] Preferably, the step of obtaining the predicted temperature change rate and the temperature change rate under the same historical working conditions for multiple time periods includes:

[0031] The current operating condition data is aligned with the historical operating condition data on the time axis, and the first-order difference sequence between the temperature data of the historical operating condition data and the temperature data under the current operating condition is calculated. The first-order difference sequence is clustered using a clustering algorithm to obtain multiple clusters, and the first-order difference means in all clusters are calculated. The time period corresponding to the cluster whose first-order difference mean is closest to 0 is the time period of stable temperature under the same operating condition after alignment.

[0032] Preferably, judging whether there is a risk in the junction box according to the change of continuous abnormal temperature values ​​under the same historical working conditions includes:

[0033] Obtain the sequence of temperature anomaly values ​​corresponding to the aligned historical data under the same working condition, and calculate the approximation between the sequence of temperature anomaly values ​​corresponding to the current collected data and the sequence of temperature anomaly values ​​corresponding to the aligned historical data under the same working condition;

[0034] The data after normalizing the temperature anomaly value is used as the approximation weighted value to obtain the risk coefficient value. In response to the risk coefficient being greater than or equal to the preset threshold, there is a risk and the control parameters of the junction box are adjusted. Otherwise, normal operation continues.

[0035] The present invention has the following effects:

[0036] 1. The present invention can identify the deviation between the actual temperature change and the expected change by comparing the current operating condition data with the historical operating condition data, thereby reducing the misjudgment caused by the abnormality of a single data point or short-term fluctuations and improving the accuracy of temperature anomaly detection.

[0037] 2. Through cluster analysis and time series data prediction, the system of the present invention can more accurately capture the trend of temperature changes, thereby reducing false alarms and missed alarms in practical applications.

[0038] 3. The present invention obtains the temperature anomaly value by calculating the temperature change rate and the temperature stability deviation, and determines whether there is a risk in the junction box by continuously monitoring the change of the temperature anomaly value, so as to take measures in advance to avoid accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0040] Figure 1 It is a method flow chart of steps S1 to S4 in a method for intelligent wiring control of a junction box according to an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0042] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0043] Reference Figure 1 , an intelligent wiring control method for a junction box includes steps S1 to S4, which are specifically as follows:

[0044] S1: Acquire the operating condition data of the sensors of the historical intelligent junction box, wherein the operating condition data includes: voltage data, current data, temperature data and timing data.

[0045] It should be noted that a temperature sensor and a voltage and current sensor are installed in the junction box to obtain time series voltage data A, current data B and time series data C, wherein the timestamps of the three sensor data are aligned, that is, within the same time period, the time series data lengths of the three data are consistent.

[0046] S2: Clustering is performed based on historical operating condition data, and historical data similar to the current operating condition data is obtained according to the clustering results, and the change rate between the current temperature value and the temperature under the same operating condition in the historical data is calculated.

[0047] Get historical data similar to the current operating data, including:

[0048] Based on the time series data, the voltage data and the current data are clustered for the first time to obtain multiple clusters, the mean power of the clusters is calculated, the clusters are sorted according to the mean power, the sorting number of each category is obtained, and the sorting numbers of all categories are normalized;

[0049] It should be noted that the first clustering cluster adopts the mean shift algorithm, where the clustering radius is 0.02. When clustering, the data of the two dimensions need to be normalized to the maximum and minimum values ​​respectively. This clustering is based on current data and voltage data, and the average power of each cluster is further calculated and sorted. The purpose is to classify the approximate current and voltage data into a cluster. The algorithm is a well-known technology in this field and will not be described in detail. The clustering algorithm can be replaced according to the specific implementation scenario;

[0050] To further explain, the first clustering is based on the preliminary classification of voltage and current data. The reason why the power is not directly calculated before clustering is that in actual operation, the voltage and current change independently, and the voltage data and current data corresponding to the same power will be different. Directly using the power value may mask abnormal voltage or current conditions. Even if the power looks normal, excessive voltage or current may be dangerous and need to be detected and processed.

[0051] The time series of sensor data is replaced with a normalized sequence of numbers, and the time points corresponding to the numbers in the sequence of numbers are clustered for the second time to obtain multiple sequence time clusters;

[0052] It should be noted that the second clustering adopts the DBSCAN algorithm, in which the clustering radius is 0.05 and the minimum number of clusters is 5, which can be adjusted by the implementer according to the specific implementation scenario. This algorithm is a public technology for personnel in this field and will not be described in detail.

[0053] It is further explained that the second clustering is based on the classification of the sequence number and time of the mean power. That is to say, clusters with close sequence numbers represent similar power, time can reflect the similarity of continuous working conditions, and the data segmentation result corresponding to the sequence number is within a time period.

[0054] Calculate the mean of the sequence number in the sequence time clustering cluster, and perform clustering for the third time to obtain multiple sequence mean clustering clusters. Get the corresponding sensor data according to the sequence number in the sequence mean clustering cluster. Calculate the approximation between the sequence number corresponding to the sequence mean clustering cluster where the current collected data is located and the sequence number data corresponding to each sequence mean clustering cluster, and obtain the maximum value of the approximation as the historical data that is similar to the current operating condition data.

[0055] It should be noted that the third clustering adopts the mean shift algorithm, in which the clustering radius is 0.02, and the clusters with similar sequence number means are further divided into one category. The purpose is to classify the segments with similar operating condition data into one category. The above approximation can be achieved by the dynamic time warping (DTW) algorithm to obtain the approximation between two sequence number data sequences, and then find the historical operating condition that is most similar to the current operating condition. In other words, the greater the approximation, the more similar the current operating condition is to the historical operating condition, and the lower the risk; it is helpful to evaluate the risk level of the current operating condition and adjust the control parameters of the smart junction box accordingly.

[0056] In addition, another implementation also includes:

[0057] The historical data and current operating condition data are converted into feature vectors respectively, the nearest neighbor search algorithm is used to search for the most similar data points, and the cosine similarity between the feature vector of the historical data and the feature vector of the current operating condition data is calculated. The searched data points of the historical data are sorted according to the cosine similarity, and the top N data points with the smallest cosine similarity are selected as the historical data that are similar to the current operating condition data.

[0058] It should be noted that the nearest neighbor search algorithm, such as KD tree, ball tree, brute force search, etc., can be selected by the implementer according to the specific implementation situation.

[0059] Get the conversion rate, including:

[0060] Align the current working condition data with the historical working condition data on the time axis, determine the starting time, calculate the time interval between the current time and the starting time of the current working condition data, find the corresponding time point in the aligned historical working condition data according to the time interval, and obtain the average temperature value at the same time point as the expected temperature;

[0061] Get the actual temperature at the time point corresponding to the current operating condition data, calculate the difference between the actual temperature and the expected temperature, and get the temperature change rate.

[0062] In addition, another embodiment further includes:

[0063] The temperature data in the current working condition data and the temperature data of multiple periods under the same working condition in history are obtained, and the temperature data of multiple periods under the same working condition in history are fitted using the least square method to obtain a fitting curve, and the average value of the same time point in the fitting curve of multiple time periods is calculated as the expected stable temperature under the same working condition in history;

[0064] Obtain the actual temperature at the same time point of the current working condition data and the expected stable temperature, calculate the difference between the actual temperature and the expected temperature, and obtain the temperature change rate.

[0065] S3: Use the time series data prediction method to predict the time series of the temperature change rate in the current operating condition data, obtain multiple time periods of the predicted temperature change rate and the temperature change rate under the same historical operating conditions, and calculate the temperature stability deviation at the same time under the same operating conditions.

[0066] Exemplarily, the time series data prediction method is data fitting or ARIMA method for prediction, which can be adjusted by the implementer according to the specific implementation scenario and will not be described in detail in this solution.

[0067] Get the predicted temperature change rate and the temperature change rate under the same historical conditions for multiple time periods, including:

[0068] The current operating condition data is aligned with the historical operating condition data on the time axis, and the first-order difference sequence between the temperature data of the historical operating condition data and the temperature data under the current operating condition is calculated. The first-order difference sequence is clustered using a clustering algorithm to obtain multiple clusters, and the first-order difference means in all clusters are calculated. The time period corresponding to the cluster whose first-order difference mean is closest to 0 is the time period of stable temperature under the same operating condition after alignment.

[0069] It should be noted that the DBSCAN algorithm is used for classification in this embodiment, where the clustering radius is 0.1 and the minimum number of clusters is 3, which can be adjusted by the implementer according to the specific implementation scenario.

[0070] Get temperature stability deviation, including:

[0071] The average value of the temperature change rate at the same time point under the same working conditions in multiple periods of history is taken as the stable temperature, and the difference between the stable temperature at the same time and the temperature change rate of the current working condition data is taken as the temperature stability deviation.

[0072] In addition, another embodiment further includes:

[0073] Align the current working condition data with the historical data on the time axis, obtain the temperature data of the historical data and the temperature data of the current working condition at the same time point, and preset a sliding window, respectively calculate the first-order difference values ​​of the temperature data in the sliding window corresponding to the current working condition data and the historical data, and obtain the first-order difference sequence of the temperature data in the window corresponding to the current working condition data and the historical data, and respectively use the mean of the first-order difference sequence of the preset sliding window as the stability of the central data point in the preset window, and when the stability is less than the stability threshold, it is used as low stable temperature data, otherwise, it is high stable temperature data, and respectively obtain the time periods corresponding to the continuous low stable temperature data in the current working condition data and the historical data, wherein there is at least one time period corresponding to the low stable temperature data;

[0074] Calculate the ratio of the temperature difference at the beginning and end of the preset sliding window for the time period corresponding to the low stable temperature data in the current working condition data and the historical data to the time interval respectively, and obtain the temperature change rate of the current working condition data and the historical data respectively. Calculate the difference between the average of the temperature change rates of multiple low stable temperature data corresponding to the time period of the current working condition data and the average of the temperature change rates of multiple low stable temperature data corresponding to the time period of the historical data respectively, and obtain the temperature stability deviation.

[0075] For example, the stability threshold is 3°C. It should be noted that the preset sliding window moves a preset interval each time. In this embodiment, the preset interval is 1, and the preset sliding window size is , implementers can make adjustments based on actual conditions until the entire data set is covered.

[0076] S4: Calculate the temperature anomaly value based on the approximate historical data and change rate of the current working condition data and the temperature stability deviation, and determine whether there is a risk in the junction box based on the change of continuous temperature anomaly values ​​under the same historical working condition.

[0077] Specifically, the temperature abnormal value Satisfies the following relationship:

[0078] ,

[0079] In the formula, Indicates historical data that is similar to the current operating data. Indicates the conversion rate between the temperature data of the current working condition data and the temperature data of the same working condition in the historical data. Indicates the temperature stability deviation between the predicted stable temperature value of the temperature value of the current working condition data and the stable temperature under the same working condition in history. Represented by natural numbers An exponential function with base .

[0080] Determine whether there is a risk in the junction box, including the following steps:

[0081] Obtain the sequence of temperature anomaly values ​​corresponding to the aligned historical data under the same working condition, and calculate the approximation between the sequence of temperature anomaly values ​​corresponding to the current collected data and the sequence of temperature anomaly values ​​corresponding to the aligned historical data under the same working condition;

[0082] The data after normalizing the temperature anomaly value is used as the approximation weighted value to obtain the risk coefficient value. In response to the risk coefficient being greater than or equal to the preset threshold, there is a risk and the control parameters of the junction box are adjusted. Otherwise, normal operation continues.

[0083] It should be noted that the approximation weighted value is calculated using weighted Pearson correlation, which is a well-known technique in the art and will not be described in detail.

[0084] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0085] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. An intelligent wiring control method for a junction box, characterized in that: include: Acquire the operating condition data of the sensor of the historical intelligent junction box, wherein the operating condition data includes: voltage data, current data, temperature data and time series data; Based on the historical working condition data, clustering is performed, and the historical data similar to the current working condition data is obtained according to the clustering results, and the change rate of the current temperature value and the temperature under the same working condition in the historical data is calculated; the temperature change rate refers to: the difference between the current temperature value and the mean value of the temperature under the same working condition in the historical data; Use the time series data prediction method to predict the time series of the temperature change rate in the current working condition data, obtain the predicted temperature change rate and the temperature change rate under the same working condition in multiple periods, and calculate the temperature stability deviation at the same time under the same working condition; Calculate the temperature anomaly value based on the historical data and change rate of the current working condition data and the temperature stability deviation. According to the change of continuous temperature anomaly values ​​under the same working condition in history, determine whether there is a risk in the junction box. Among them, the temperature abnormal value Satisfies the following relationship: , where Indicates historical data that is similar to the current operating data. Indicates the conversion rate between the temperature data of the current working condition data and the temperature data of the same working condition in the historical data. Indicates the temperature stability deviation between the predicted stable temperature value of the temperature value of the current working condition data and the stable temperature value under the same working condition in history. Represented by natural numbers An exponential function with base .

2. The intelligent wiring control method of a junction box according to claim 1, characterized in that: Obtaining historical data that is similar to the current operating condition data, including: performing a first clustering of voltage data and current data based on time series data to obtain multiple clustering clusters, calculating the mean power of the clustering clusters, sorting the clustering clusters according to the mean power, obtaining the sorting sequence number of each category, and performing normalization processing; replacing the time series sequence of the sensor data with the normalized sequence sequence, performing a second clustering of the time points corresponding to the sequence numbers in the sequence sequence, and obtaining multiple sequence time clustering clusters; calculating the mean of the sequence numbers in the sequence time clustering clusters, and performing a third clustering to obtain multiple sequence mean clustering clusters, obtaining the corresponding sensor data according to the sequence numbers in the sequence mean clustering clusters, calculating the approximation between the sequence number corresponding to the sequence mean clustering cluster where the current collected data is located and the sequence number data corresponding to each sequence mean clustering cluster, and obtaining the maximum value of the approximation as the historical data that is similar to the current operating condition data.

3. The intelligent wiring control method of a junction box according to claim 1, characterized in that: Obtain historical data similar to the current operating data, including: The historical data and current operating condition data are converted into feature vectors respectively, the nearest neighbor search algorithm is used to search for the most similar data points, and the cosine similarity between the feature vector of the historical data and the feature vector of the current operating condition data is calculated. The searched data points of the historical data are sorted according to the cosine similarity, and the top N data points with the smallest cosine similarity are selected as the historical data that are similar to the current operating condition data.

4. The intelligent wiring control method of a junction box according to claim 1, characterized in that: The conversion rate includes: Align the current working condition data with the historical working condition data on the time axis, determine the starting time, calculate the time interval between the current time and the starting time of the current working condition data, find the corresponding time point in the aligned historical working condition data according to the time interval, and obtain the average temperature value at the same time point as the expected temperature; Get the actual temperature at the time point corresponding to the current operating condition data, calculate the difference between the actual temperature and the expected temperature, and get the temperature change rate.

5. The intelligent wiring control method of a junction box according to claim 1, characterized in that: The conversion rate also includes: The temperature data in the current working condition data and the temperature data of multiple periods under the same working condition in history are obtained, and the temperature data of multiple periods under the same working condition in history are fitted using the least square method to obtain a fitting curve, and the average value of the same time point in the fitting curve of multiple time periods is calculated as the expected stable temperature under the same working condition in history; Obtain the actual temperature at the same time point of the current working condition data and the expected stable temperature, calculate the difference between the actual temperature and the expected temperature, and obtain the temperature change rate.

6. The intelligent wiring control method of a junction box according to claim 1, characterized in that: The temperature stability deviation includes: The average value of the temperature change rate at the same time point under the same working conditions in multiple periods of history is taken as the stable temperature, and the difference between the stable temperature at the same time and the temperature change rate of the current working condition data is taken as the temperature stability deviation.

7. The intelligent wiring control method of a junction box according to claim 1, characterized in that: The temperature stability deviation also includes: Align the current working condition data with the historical data on the time axis, obtain the temperature data of the historical data and the temperature data of the current working condition at the same time point, and preset a sliding window, respectively calculate the first-order difference values ​​of the temperature data in the sliding window corresponding to the current working condition data and the historical data, and obtain the first-order difference sequence of the temperature data in the window corresponding to the current working condition data and the historical data, and respectively use the mean of the first-order difference sequence of the preset sliding window as the stability of the central data point in the preset window, and when the stability is less than the stability threshold, it is used as low stable temperature data, otherwise, it is high stable temperature data, and respectively obtain the time periods corresponding to the continuous low stable temperature data in the current working condition data and the historical data, wherein there is at least one time period corresponding to the low stable temperature data; Calculate the ratio of the temperature difference at the beginning and end of the preset sliding window of the time period corresponding to the low stable temperature data in the current working condition data and the historical data to the time interval respectively, and obtain the temperature change rate of the current working condition data and the historical data respectively. Calculate the difference between the average of the temperature change rates of multiple low stable temperature data corresponding to the time period of the current working condition data and the average of the temperature change rates of multiple low stable temperature data corresponding to the time period of the historical data respectively, and obtain the temperature stability deviation.

8. The intelligent wiring control method of a junction box according to claim 1, characterized in that: The method of obtaining the predicted temperature change rate and the temperature change rate under the same historical working conditions for multiple time periods includes: The current operating condition data is aligned with the historical operating condition data on the time axis, and the first-order difference sequence between the temperature data of the historical operating condition data and the temperature data under the current operating condition is calculated. The first-order difference sequence is clustered using a clustering algorithm to obtain multiple clusters, and the first-order difference means in all clusters are calculated. The time period corresponding to the cluster whose first-order difference mean is closest to 0 is the time period of stable temperature under the same operating condition after alignment.

9. The intelligent wiring control method of a junction box according to claim 1, characterized in that: The determination of whether there is a risk in the junction box based on the change of continuous abnormal temperature values ​​under the same historical working conditions includes: Obtain the sequence of temperature anomaly values ​​corresponding to the aligned historical data under the same working condition, and calculate the approximation between the sequence of temperature anomaly values ​​corresponding to the current collected data and the sequence of temperature anomaly values ​​corresponding to the aligned historical data under the same working condition; The data after normalizing the temperature anomaly values ​​is used as the approximation weighted value to obtain the risk coefficient value. If the risk coefficient is greater than or equal to the preset threshold, there is a risk and the control parameters of the junction box are adjusted. Otherwise, normal operation continues.

Images