Cloud platform-based distributed energy data monitoring method, device, equipment and medium

Abstract

The application discloses a cloud platform-based distributed energy data monitoring method and device, equipment and medium, and belongs to the technical field of cloud computing, which comprises the following steps: collecting monitoring operation data of an energy node monitoring device in real time to obtain an operation data set; calculating an operation state evaluation value of the monitoring device according to the operation data set; judging the size of the operation state evaluation value and a preset threshold value; selecting a real-time monitoring mode, a mixed monitoring mode and a prediction monitoring mode according to the judgment result, and outputting monitoring data by using the selected mode. Through the optimization of the multi-mode monitoring and the prediction model, the application realizes real-time, accurate and efficient monitoring of the operation state of the energy device.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cloud computing, in particular to a distributed energy data monitoring method, device and equipment based on a cloud platform and a medium. BACKGROUND

[0002] With the increasing demand for energy and the growing environmental problems, distributed energy systems have developed rapidly. Distributed energy systems have characteristics such as decentralization, diversity and dynamics, which put higher requirements on data monitoring technology.

[0003] At present, the monitoring technology of distributed energy systems mainly relies on traditional real-time data acquisition and analysis methods. Real-time acquisition of a large amount of data causes resource waste and bandwidth pressure, and real-time monitoring is redundant for devices with stable running state. Secondly, when the monitoring device is abnormal, the response speed of real-time monitoring may not be sufficient, and the problem cannot be discovered and handled in time, leading to accumulation of potential risks. Finally, the traditional monitoring system lacks intelligent operation and maintenance strategies, and it is difficult to flexibly adjust the monitoring mode according to the device running state, resulting in low monitoring efficiency. SUMMARY

[0004] To solve the above problems, the present application provides a distributed energy data monitoring method, device and equipment based on a cloud platform and a medium, which realizes real-time, accurate and efficient monitoring of the running state of energy equipment through multi-mode monitoring and optimization of prediction models.

[0005] The above-mentioned object can be achieved by the following scheme:

[0006] The distributed energy data monitoring method based on a cloud platform comprises: collecting real-time monitoring running data of energy node monitoring devices to obtain a running data set; calculating the running state evaluation value of the monitoring device according to the running data set; judging the size of the running state evaluation value and the preset threshold value; selecting real-time monitoring mode, hybrid monitoring mode and prediction monitoring mode according to the judgment result, and outputting monitoring data by using the selected mode.

[0007] Further, the calculation of the running state evaluation value of the monitoring device according to the running data set comprises: obtaining historical monitoring running data and historical running state evaluation value of the monitoring device from a preset database to obtain a historical data set; establishing a running state evaluation function for representing the running state evaluation value; optimizing the parameters of the running state evaluation function by using the historical data set to obtain the final running state evaluation function; inputting the running data set into the final running state evaluation function to obtain the running state evaluation value, and for the running state evaluation value S, there is:

[0008]

[0009] In the formula, a i is a weight coefficient of the i-th monitoring operation data in the operation data set, B i is the i-th monitoring operation data in the operation data set, C is an error coefficient, and n is an integer greater than 0.

[0010] Further, the judging the size of the operation state evaluation value and the preset threshold value comprises: judging whether the operation state evaluation value is greater than a preset first threshold value; if the operation state evaluation value is greater than the first threshold value, selecting a real-time monitoring mode, performing energy data monitoring and outputting monitoring data; if the operation state evaluation value is less than or equal to the first threshold value, judging whether the operation state evaluation value is greater than a preset second threshold value; if the operation state evaluation value is greater than the second threshold value, selecting a hybrid monitoring mode, performing energy data monitoring and outputting monitoring data; and if the operation state evaluation value is less than or equal to the second threshold value, selecting a prediction monitoring mode, performing energy data monitoring and outputting monitoring data.

[0011] Further, the selecting the real-time monitoring mode, performing energy data monitoring and outputting monitoring data comprises: collecting data of the energy node in real time; performing feature extraction on the collected data to obtain monitoring data; and outputting the monitoring data.

[0012] Further, the selecting the prediction monitoring mode, performing energy data monitoring and outputting monitoring data further comprises: collecting historical monitoring data of the energy node to obtain a training data set; using the monitoring data of a previous moment as input and the monitoring data of a current moment as output, constructing and training a neural network model by using the training data set to obtain a monitoring data prediction model; obtaining the monitoring data of the previous moment and inputting the monitoring data prediction model to obtain current monitoring prediction data; and outputting the current monitoring prediction data.

[0013] Further, the hybrid monitoring mode, performing energy data monitoring and outputting monitoring data comprises: collecting data of the energy node in real time and judging whether the collected data is abnormal; if the collected data is not abnormal, performing feature extraction on the collected data to obtain current monitoring data and outputting the current monitoring data; and if the collected data is abnormal, inputting the monitoring data of the previous moment into the monitoring data prediction model to obtain current monitoring prediction data and outputting the current monitoring prediction data.

[0014] Further, the method further comprises: collecting data of the energy node in real time and performing feature extraction to obtain actual monitoring data; outputting the actual monitoring data and sending an early warning information when an error between the actual monitoring data and current monitoring prediction data is greater than a preset condition; checking the monitoring device according to the early warning information and feeding back a checking result; if the checking result is abnormal, the monitoring data prediction model does not need to be optimized; if the checking result is normal, the monitoring data prediction model is optimized according to the actual monitoring data.

[0015] Based on the same inventive concept, the application further provides a cloud platform-based distributed energy data monitoring device, which comprises: a data collection module, configured to collect monitoring operation data of an energy node monitoring device in real time to obtain an operation data set; an evaluation value calculation module, configured to calculate an operation state evaluation value of the monitoring device according to the operation data set; an evaluation value analysis module, configured to judge the size of the operation state evaluation value and a preset threshold; a monitoring mode selection module, configured to select a real-time monitoring mode, a hybrid monitoring mode and a prediction monitoring mode according to the judgment result, and output monitoring data by using the selected mode.

[0016] Based on the same inventive concept, the application further provides a computer storage medium, which stores one or more programs, and when the one or more programs are executed, the method described above can be implemented.

[0017] Based on the same inventive concept, the application further provides a device, which comprises a processor, a communication interface, a memory and a communication bus; the processor, the communication interface and the memory communicate with each other through the communication bus; the processor is configured to execute the program stored in the computer readable storage medium described above.

[0018] Compared with the prior art, the application has the following advantages:

[0019] 1. The application dynamically selects a real-time monitoring mode, a hybrid monitoring mode and a prediction monitoring mode according to an operation state evaluation value by collecting operation data of an energy node monitoring device in real time; the intelligent monitoring mode selection mechanism can be adjusted according to the actual operation state of the monitoring device, thereby improving the monitoring efficiency and avoiding the problems of resource waste and monitoring blind area;

[0020] 2. The application realizes efficient and accurate monitoring of energy data by combining the advantages of real-time data collection and a prediction model when the operation state of the monitoring device is general, effectively avoids monitoring interruption or error data output caused by data abnormalities, and improves the stability and reliability of the monitoring system;

[0021] 3、The application can reduce the dependence on real-time monitoring by predicting the current monitoring data when the running state of the monitoring equipment is poor, which helps to improve the response speed of the system when the running state of the monitoring equipment is poor, ensures that measures can be taken quickly when the equipment fails, reduces the impact of the failure on the system, and improves the stability and reliability of the system;

[0022] 4、The application also includes real-time data verification and model optimization processes, which can discover and correct errors in time by comparing actual monitoring data with predicted data, and ensure the accuracy of the data; at the same time, the prediction model is optimized according to the actual monitoring data, which improves the prediction ability and adaptability of the model;

[0023] 5、The technical scheme of the application is based on a cloud platform, which has the advantages of easy deployment and maintenance. This reduces the operation and maintenance cost of the system, improves the scalability and flexibility of the system, and makes it better adapt to distributed energy systems of different scales and needs.

[0024] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings. BRIEF DESCRIPTION OF DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0026] Figure 1 is a flowchart of the cloud platform-based distributed energy data monitoring method of the embodiment of the present application.

[0027] Figure 2 is an execution flowchart of the cloud platform-based distributed energy data monitoring method of the embodiment of the present application.

[0028] Figure 3 is an execution flowchart of the real-time monitoring mode of the embodiment of the present application.

[0029] Figure 4 is an execution flowchart of the prediction monitoring mode of the embodiment of the present application.

[0030] Figure 5 is an execution flowchart of the hybrid monitoring mode of the embodiment of the present application.

[0031] Figure 6is a structural schematic diagram of a cloud platform-based distributed energy data monitoring device according to an embodiment of the present application.

[0032] Figure 7 is a structural schematic diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0033] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some, but not all, of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts should fall within the scope of the present application.

[0034] With reference to Figure 1 An embodiment of the present application proposes a cloud platform-based distributed energy data monitoring method, which realizes real-time, accurate, and efficient monitoring of the running state of energy equipment through multi-mode monitoring and optimization of prediction models.

[0035] The method of this embodiment specifically includes:

[0036] Real-time collection of monitoring running data of the energy node monitoring device to obtain a running data set;

[0037] Specifically, real-time acquisition of running data from the monitoring devices of each energy node (such as solar panels, wind turbines, energy storage devices, etc.) of the distributed energy system; these data can include current, voltage, power, temperature, humidity, and other parameters, depending on the type of monitoring device and the type of energy being monitored.

[0038] According to the running data set, the running state evaluation value of the monitoring device is calculated;

[0039] Specifically, the running state of the monitoring device is evaluated using the running data set; this usually requires an evaluation function or model that can calculate the running state evaluation value of the device according to real-time running data; this evaluation value can reflect the health status, performance level, or potential failure risk of the device.

[0040] Determine the size of the running state evaluation value and the preset threshold value;

[0041] According to the judgment result, select the real-time monitoring mode, the mixed monitoring mode, and the prediction monitoring mode, and output the monitoring data using the selected mode.

[0042] Specifically, according to the comparison result of the running state evaluation value and the preset threshold value, a suitable monitoring mode is selected to output the monitoring data; the real-time monitoring mode involves real-time data collection and output; the hybrid monitoring mode combines the characteristics of real-time monitoring and predictive monitoring; the predictive monitoring mode mainly relies on the prediction model to output data; the real-time, accurate and efficient monitoring of the running state of the energy equipment can be realized, thereby improving the stability and reliability of the system.

[0043] Further, according to the running data set, the running state evaluation value of the monitoring device is calculated, including:

[0044] The historical monitoring running data and the historical running state evaluation value of the monitoring device are obtained from the preset database to obtain a historical data set;

[0045] Specifically, the historical monitoring running data of the same or similar device as the current monitoring device and the corresponding historical running state evaluation value are extracted from the preset database; these data are used to train or optimize the running state evaluation function to improve its accuracy and reliability; the historical data set should contain data under various running states to fully reflect the performance characteristics of the device.

[0046] A running state evaluation function for representing the running state evaluation value is established;

[0047] Specifically, according to the characteristics and monitoring requirements of the monitoring device, a function capable of representing the running state evaluation value is established; this function can be a mathematical formula, a machine learning model or a deep learning network, etc., and the specific form depends on the characteristics of the data and the complexity of the problem; the purpose of the function is to calculate the running state evaluation value of the device according to the real-time running data.

[0048] The parameters of the running state evaluation function are optimized using the historical data set to obtain the final running state evaluation function;

[0049] Specifically, for the running state evaluation function based on multiple linear regression, optimization algorithms such as gradient descent or least squares can be used to adjust the weight coefficients; through iterative calculation, a set of optimal weight coefficients can be found, so that the error between the predicted health index and the actual value is minimized; this set of weight coefficients will be used in the final running state evaluation function.

[0050] The running data set is input into the final running state evaluation function to obtain the running state evaluation value, and for the running state evaluation value S, there is:

[0051]

[0052] In the formula, a i is the weight coefficient of the i-th monitoring running data in the running data set, B iCi is the error coefficient, and n is an integer greater than 0.

[0053] Exemplarily, it is assumed that the final operation state evaluation function has been obtained, and a set of monitoring operation data (for example, the wind speed can be 12 m / s, the power output can be 2 MW, and the temperature can be 20°C) is collected in real time; the operation state evaluation value of the device is obtained by inputting the data into the operation state evaluation function. According to the operation state evaluation value, it can be judged whether the operation state of the device is good, and a suitable monitoring mode is selected to output monitoring data; through the above process, accurate evaluation of the operation state of the energy device can be realized, and strong support is provided for the selection of the subsequent monitoring mode.

[0054] Further, as shown in Figure 2 , judging the size of the operation state evaluation value and the preset threshold value comprises:

[0055] judging whether the operation state evaluation value is greater than a preset first threshold value;

[0056] If the operation state evaluation value is greater than the first threshold value, a real-time monitoring mode is selected, energy data monitoring is performed, and monitoring data is output;

[0057] If the operation state evaluation value is less than or equal to the first threshold value, it is judged whether the operation state evaluation value is greater than a preset second threshold value;

[0058] If the operation state evaluation value is greater than the second threshold value, a hybrid monitoring mode is selected, energy data monitoring is performed, and monitoring data is output;

[0059] If the operation state evaluation value is less than or equal to the second threshold value, a prediction monitoring mode is selected, energy data monitoring is performed, and monitoring data is output.

[0060] Exemplarily, it is assumed that there is a monitoring device, and the operation state evaluation value of the monitoring device calculated by the operation state evaluation function is 80; it is assumed that the first threshold value is 70 and the second threshold value is 40, and since the operation state evaluation value 80 is greater than the first threshold value 70, a real-time monitoring mode is selected for energy data monitoring and monitoring data is output; through the above judgment process and monitoring mode selection mechanism, the distributed energy data monitoring system based on the cloud platform can flexibly adjust the monitoring strategy according to the actual operation state of the device, and realize efficient and accurate monitoring.

[0061] Further, as shown in Figure 3 , selecting a real-time monitoring mode, performing energy data monitoring, and outputting monitoring data comprises:

[0062] collecting data of the energy node in real time;

[0063] Specifically, in real-time monitoring mode, data is continuously and uninterruptedly collected from energy nodes; the collected data can include voltage, current, power, frequency, temperature, humidity, and other parameters, depending on the type of equipment and monitoring requirements; the collection process is usually implemented through sensors, data collectors, and other devices, which convert analog signals into digital signals for subsequent processing and analysis.

[0064] Feature extraction is performed on the collected data to obtain monitoring data.

[0065] Specifically, feature extraction is one of the key steps in data processing. In real-time monitoring mode, the collected raw data is processed and analyzed to extract meaningful features for monitoring. Feature extraction can include data cleaning, denoising, normalization, transformation, and calculation of statistical quantities, trend analysis, and anomaly detection. The extracted features will be used as monitoring data for subsequent judgment, decision-making, and output.

[0066] The monitoring data is output.

[0067] Specifically, outputting monitoring data is one of the final goals of the monitoring system. In real-time monitoring mode, the extracted monitoring data is presented to the user in a visual manner or transmitted to other systems through an interface. The output data can include real-time curves, reports, alerts, and other information to help users understand the running status and performance of the equipment in a timely manner.

[0068] Further, as shown in Figure 4 selecting a prediction monitoring mode to monitor energy data and output monitoring data further includes:

[0069] Collecting historical monitoring data of energy nodes to obtain a training data set;

[0070] Using the training data set to build and train a neural network model, with the monitoring data of the previous time as input and the monitoring data of the current time as output, to obtain a monitoring data prediction model;

[0071] Specifically, in the prediction monitoring mode, first, the historical monitoring data of the energy node needs to be collected as the training data set for training the neural network model; the historical monitoring data should contain data of multiple time periods so that the model can learn the change rule and trend of the data; the data collection process is usually realized through data collectors, databases, etc. to ensure the accuracy and integrity of the data; after collecting the training data set, the neural network model needs to be constructed and trained using these data; the input of the model is the monitoring data at the previous moment, and the output is the monitoring data at the current moment; the training process usually includes data preprocessing, model construction, parameter optimization, etc. to enable the model to accurately predict future monitoring data; common neural network models include feedforward neural networks, convolutional neural networks, recurrent neural networks, etc. The specific choice depends on the characteristics of the data and the prediction requirements.

[0072] For example, assuming that a photovoltaic power generation system is being monitored, historical monitoring data such as power generation, light intensity, temperature, etc. in the past year needs to be collected; these data can be extracted from the data collection system of the photovoltaic power station and stored in the database as the training data set; for the photovoltaic power generation system, a feedforward neural network can be selected as the prediction model, the input of the model is the monitoring data such as power generation, light intensity, temperature, etc. at the previous moment, and the output is the predicted value of power generation at the current moment; the model is trained through the training data set to adjust the weight and bias parameters of the model so that the model can accurately predict future power generation.

[0073] Obtain the monitoring data at the previous moment and input it into the monitoring data prediction model to obtain the current monitoring prediction data;

[0074] Output the current monitoring prediction data.

[0075] Specifically, after the monitoring data prediction model is trained, the monitoring data at the previous moment needs to be obtained and input into the model for prediction, and the model will calculate the current monitoring prediction data according to the input data; after obtaining the current monitoring prediction data, it needs to be output to the user interface or other system so that the user can timely understand the running state and performance of the device; the output data can include predicted value, prediction interval, prediction confidence, etc. to enable the user to make decisions and judgments.

[0076] Further, as shown in Figure 5 the mixed monitoring mode, monitoring energy data and outputting monitoring data include:

[0077] Real-time acquisition of data of the energy node and judgment of whether the acquired data is abnormal;

[0078] Specifically, in the hybrid monitoring mode, the data of the energy node will be collected in real time first, which may include key parameters such as voltage, current, power, temperature, etc. After the data is collected, the data will be analyzed immediately to determine whether they are within the normal range, i.e. whether there is an anomaly. The anomaly determination may be based on a pre-set threshold, statistical rules of historical data or machine learning algorithms, etc.

[0079] If the collected data is not abnormal, the collected data will be feature extracted to obtain the current monitoring data and output.

[0080] Specifically, if the real-time collected data is not abnormal, the data will be feature extracted, such as calculating the average, maximum, minimum, volatility, etc. to obtain the current monitoring data. The extracted monitoring data will be output to the user interface or other systems so that the user can understand the running state of the device in time.

[0081] If the collected data is abnormal, the monitoring data of the previous moment will be input into the monitoring data prediction model to obtain the current monitoring prediction data and output.

[0082] Specifically, if the real-time collected data is abnormal, the abnormal data will not be output directly, but the previously trained monitoring data prediction model will be used for prediction. The monitoring data of the previous moment will be input into the prediction model, and the model will predict the current monitoring data according to the historical data and learned rules. The predicted monitoring data will be output to the user interface or other systems to replace the abnormal real-time data, ensuring the continuity and accuracy of the monitoring.

[0083] Illustratively, suppose a wind power system is being monitored, the system will collect data such as wind speed, rotor speed, generator temperature, etc. in real time. If the wind speed suddenly drops to a value far below the normal range, or the generator temperature rises sharply, it will be determined that the data is abnormal. For a wind power system, if the real-time collected wind speed, rotor speed, generator temperature, etc. are within the normal range, the average and volatility of these data will be calculated as the current monitoring data output. If the real-time collected wind speed data suddenly drops abnormally, the wind speed, rotor speed, generator temperature, etc. of the previous moment will be input into the prediction model. The model will predict the current wind speed value according to these historical data and learned wind speed change rules, and output it as monitoring data. The hybrid monitoring mode combines the advantages of real-time monitoring and prediction monitoring by collecting data in real time and judging its abnormality, which can not only understand the current state of the device in time, but also provide accurate prediction data when the data is abnormal, ensuring the continuity and accuracy of the monitoring. This mode has wide application prospects in distributed energy data monitoring systems.

[0084] Further, as shown in Figure 4 the method further comprises:

[0085] Real-time data acquisition of the energy node and feature extraction to obtain actual monitoring data;

[0086] Specifically, the data of the energy node is continuously and real-time collected, which may include key parameters such as power, voltage, current, temperature, and pressure; the collected data is processed by feature extraction, such as calculating the average value, peak value, and fluctuation rate, to obtain actual monitoring data.

[0087] When the error between the actual monitoring data and the current monitoring prediction data is greater than the preset condition, the actual monitoring data is output and a warning information is sent;

[0088] Specifically, the actual monitoring data is compared with the prediction data output by the monitoring data prediction model, and if the error exceeds the preset threshold or condition, it is considered that the prediction model may be inaccurate, so the actual monitoring data is output and a warning information is sent to the operation and maintenance personnel.

[0089] According to the warning information, the monitoring equipment is checked and the checking result is fed back;

[0090] Specifically, after receiving the warning information, the operation and maintenance personnel will check the related monitoring equipment to confirm whether there is an abnormality; the checking result will be fed back to the system for subsequent judgment and decision-making.

[0091] If the checking result is abnormal, the monitoring data prediction model does not need to be optimized;

[0092] Specifically, if the checking result shows that the monitoring equipment itself has an abnormality, the error of the prediction model may be caused by the equipment failure rather than the problem of the model itself; in this case, the monitoring data prediction model does not need to be optimized, but the faulty equipment should be repaired or replaced.

[0093] If the checking result is normal, the monitoring data prediction model is optimized according to the actual monitoring data.

[0094] Specifically, as shown in Figure 5 if the checking result shows that the monitoring equipment is working normally, the error of the prediction model may be caused by the inaccuracy or obsolescence of the model itself; in this case, the monitoring data prediction model is optimized according to the actual monitoring data to improve the accuracy and reliability of the model.

[0095] Exemplarily, in an intelligent power grid system, the voltage, current, power factor and other data of a transformer substation are collected in real time, and the average value and fluctuation rate of the data are calculated as actual monitoring data; if the error between the actual voltage value and the predicted voltage value exceeds ±5%, the actual voltage value is output, and a warning information is sent to prompt the operation and maintenance personnel to check the voltage monitoring equipment or the prediction model; after receiving the voltage abnormality warning information, the operation and maintenance personnel go to the transformer substation to check the voltage monitoring equipment and confirm whether the equipment is normal, damaged or faulty; the checking result (such as normal equipment, damaged equipment, etc.) is fed back to the system; if the operation and maintenance personnel find that the voltage monitoring equipment is damaged, the voltage prediction model will not be optimized, but the equipment will be repaired or replaced; if the operation and maintenance personnel confirm that the voltage monitoring equipment is normal, the system will use the actual voltage data to train and optimize the voltage prediction model to improve the prediction accuracy of the model; this method ensures the accuracy and reliability of the monitoring system by collecting data in real time, comparing the error between the actual monitoring data and the predicted data, sending a warning information, checking the monitoring equipment and feeding back the checking result, and optimizing the prediction model according to the checking result. This has important application value in the distributed energy data monitoring system.

[0096] Based on the same inventive concept, as Figure 6 shown, the present application also provides a cloud platform-based distributed energy data monitoring device, which comprises:

[0097] A data collection module is configured to collect monitoring operation data of the energy node monitoring equipment in real time to obtain an operation data set.

[0098] An evaluation value calculation module is configured to calculate an operation state evaluation value of the monitoring equipment according to the operation data set.

[0099] An evaluation value analysis module is configured to judge the size of the operation state evaluation value and a preset threshold value.

[0100] A monitoring mode selection module is configured to select a real-time monitoring mode, a hybrid monitoring mode and a prediction monitoring mode according to the judgment result, and output monitoring data by using the selected mode.

[0101] Based on the above disclosure, the present application also provides an electronic device. As Figure 7 shown, the electronic device of the present application comprises at least one processor and at least one storage medium electrically connected, and the storage medium is electrically connected with the processor, wherein the storage medium stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method as above.

[0102] Based on the same inventive concept, the application further provides a storage medium, which stores instructions executable by at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method.

[0103] It should be noted that the electrical connection between the above-mentioned various units does not necessarily mean the direct connection of the line, the indirect connection mode, as long as the purpose of the application is achieved, which is applicable to the embodiments of the application. The above is only an exemplary embodiment of the application, and cannot limit the scope of the application.

[0104] That is, any equivalent changes and modifications made in accordance with the teachings of the present application are still within the scope of the present application. Other embodiments of the application will be readily apparent to those skilled in the art upon considering the description and practice of the disclosure. The present application is intended to cover any variations, uses, or adaptive changes of the application that follow the general principles of the application and include common knowledge or conventional techniques in the art not described in the application.

Claims

1. A cloud platform-based distributed energy data monitoring method, characterized in that, The method comprises: collecting monitoring operation data of the energy node monitoring device in real time to obtain an operation data set; According to the operation data set, an operation state evaluation value of the monitoring device is calculated; The size of the operation state evaluation value and a preset threshold value is judged; According to the judgment result, a real-time monitoring mode, a hybrid monitoring mode and a prediction monitoring mode are selected, and monitoring data is output using the selected mode; The judgment of the size of the operation state evaluation value and the preset threshold value comprises: judging whether the operation state evaluation value is greater than a preset first threshold value; If the operation state evaluation value is greater than the first threshold value, the real-time monitoring mode is selected, energy data monitoring is performed and monitoring data is output; If the operation state evaluation value is less than or equal to the first threshold value, it is judged whether the operation state evaluation value is greater than a preset second threshold value; If the operation state evaluation value is greater than the second threshold value, the hybrid monitoring mode is selected, energy data monitoring is performed and monitoring data is output; If the operation state evaluation value is less than or equal to the second threshold value, the prediction monitoring mode is selected, energy data monitoring is performed and monitoring data is output; The selection of the prediction monitoring mode, the energy data monitoring and the output of the monitoring data further comprise: collecting historical monitoring data of the energy node to obtain a training data set; Using the training data set, a neural network model is constructed and trained with the previous time monitoring data as input and the current time monitoring data as output to obtain a monitoring data prediction model; The monitoring data of the previous moment is obtained and input into the monitoring data prediction model to obtain the current monitoring prediction data; The current monitoring prediction data is output; The hybrid monitoring mode, the energy data monitoring and the output of the monitoring data comprise: real-time collection of data of the energy node and judgment of whether the collected data is abnormal; If the collected data is not abnormal, feature extraction is performed on the collected data, the current monitoring data is obtained and output; If the collected data is abnormal, the monitoring data of the previous moment is input into the monitoring data prediction model, the current monitoring prediction data is obtained and output. 2.The cloud platform based distributed energy data monitoring method of claim 1, wherein, The calculation of the operation state evaluation value of the monitoring device according to the operation data set comprises: obtaining historical monitoring operation data and historical operation state evaluation values of the monitoring device from a preset database to obtain a historical data set; An operation state evaluation function for representing the operation state evaluation value is established; The parameters of the operation state evaluation function are optimized using the historical data set to obtain a final operation state evaluation function; The operation data set is input into the final operation state evaluation function to obtain the operation state evaluation value, and for the operation state evaluation value S, there is: ， In the formula, ai is a weight coefficient of the i th monitoring operation data in the operation data set, Bi is the i th monitoring operation data in the operation data set, C is an error coefficient, and n is an integer greater than 0. 3.The cloud platform based distributed energy data monitoring method of claim 1, wherein, The selection of the real-time monitoring mode, the energy data monitoring and the output of the monitoring data comprise: real-time collection of data of the energy node; Feature extraction is performed on the collected data to obtain monitoring data; The monitoring data is output.

4. The cloud platform based distributed energy data monitoring method according to claim 3, wherein, The method further comprises: collecting data of the energy node in real time and performing feature extraction to obtain actual monitoring data; when the error between the actual monitoring data and the current monitoring prediction data is greater than a preset condition, outputting the actual monitoring data and sending an early warning information; checking the monitoring device according to the early warning information and feeding back a checking result; if the checking result is abnormal, the monitoring data prediction model does not need to be optimized; if the checking result is normal, the monitoring data prediction model is optimized according to the actual monitoring data.

5. A cloud platform based distributed energy data monitoring device for implementing the cloud platform based distributed energy data monitoring method according to any one of claims 1-4, characterized in that, The device comprises: a data collection module, configured to collect monitoring operation data of an energy node monitoring device in real time to obtain an operation data set; an evaluation value calculation module, configured to calculate an operation state evaluation value of the monitoring device according to the operation data set; an evaluation value analysis module, configured to judge the size of the operation state evaluation value and a preset threshold value; a monitoring mode selection module, configured to select a real-time monitoring mode, a hybrid monitoring mode and a prediction monitoring mode according to a judgment result, and output monitoring data by using the selected mode.

6. A computer storage medium, characterized in that One or more programs are stored, and when the one or more programs are executed, the cloud platform-based distributed energy data monitoring method in any one of claims 1-4 can be implemented.

7. An apparatus, comprising: The device comprises a processor, a communication interface, a memory and a communication bus; the memory stores at least one program which can be loaded and executed by the processor and is in the computer storage medium in claim 6.

Images