Power station early warning method based on historical data, electronic equipment and system

Abstract

The invention provides a historical data-based power station early warning method, electronic equipment and a historical data-based power station early warning system, and the method comprises the steps: obtaining thermal images shot by a generator set at each time point, and determining a load intensity value corresponding to each thermal image; calculating a load mean value based on the load intensity values corresponding to a plurality of continuous time points before the current time point, and judging whether to carry out early warning or not based on the load mean value; if it is judged that early warning is not carried out based on the load mean value, the load mean value is recorded, a load fluctuation vector is constructed based on the load intensity values corresponding to a plurality of continuous time points before the current time point, and a load fluctuation value is calculated based on the load fluctuation vector; and calculating a load loss value based on the load fluctuation value and the load mean value, and judging whether to carry out early warning or not based on the load loss value. According to the scheme, a two-stage processing mode is adopted, data of the sensor can be further mined and calculated, and the timeliness of early warning is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power plant management, and in particular to a power plant early warning method, electronic equipment and system based on historical data. Background Art

[0002] As the core facility of energy supply in modern society, the safe and stable operation of power generation equipment in power plants is directly related to the reliability of power supply and the smooth operation of the national economy. Therefore, it is extremely important to conduct safety early warning for power generation equipment in power plants.

[0003] First of all, power generation equipment is the heart of a power plant. Once a failure or accident occurs, it will not only cause power supply interruption, but may also cause serious consequences such as fire and explosion, posing a huge threat to the safety of life and property. Through the safety early warning system, the operating status of the equipment can be monitored in real time, and potential safety hazards can be discovered in time, so as to take preventive measures to avoid accidents. Secondly, power generation equipment usually operates in harsh environments such as high temperature, high pressure, and strong electromagnetic fields. These environmental factors will accelerate the aging and damage of the equipment. The safety early warning system can predict the life and maintenance cycle of the equipment by analyzing the equipment operation data, provide a scientific basis for the maintenance and management of the equipment, extend the service life of the equipment, and reduce maintenance costs.

[0004] In summary, safety early warning of power plant power generation equipment is an important means to ensure safe, stable and reliable power supply. It can not only prevent accidents, protect the lives and property of personnel, but also improve the operating efficiency and economic benefits of equipment, and contribute to the sustainable development of the power industry. Therefore, power plants should attach great importance to the safety early warning of power generation equipment, continuously improve the early warning mechanism, enhance the early warning capability, and ensure the safe and stable operation of power plants. However, existing technologies often rely only on the original monitoring data of sensor equipment for early warning, such as temperature and pressure, which is often not timely enough.

[0005] In view of this, the present invention is proposed. Summary of the invention

[0006] The purpose of the present invention is to provide a power station early warning method, electronic equipment and system based on historical data. This scheme is a two-stage processing method. In the first stage of processing, an early warning is issued for a continuous high load state; in the second stage of processing, it is determined whether to issue an early warning based on the load mean and load fluctuation. This scheme can further mine and calculate the sensor data to improve the timeliness of the early warning.

[0007] The present invention provides a power plant early warning method based on historical data, the method comprising the following steps:

[0008] Obtain thermal images taken of the generator set at each time point, and determine the load intensity value corresponding to each thermal image;

[0009] Calculate the load average based on the load intensity values ​​corresponding to multiple consecutive time points before the current time point, and determine whether to issue an early warning based on the load average;

[0010] If it is determined that no warning is issued based on the load average, the load average is recorded, and a load fluctuation vector is constructed based on the load intensity values ​​corresponding to multiple consecutive time points before the current time point, and the load fluctuation value is calculated based on the load fluctuation vector;

[0011] A load loss value is calculated based on the load fluctuation value and the load average value, and whether to issue an early warning is determined based on the load loss value.

[0012] The above scheme is adopted. This scheme is a two-stage processing method. As the load of the generator set increases, the temperature of each position of the generator set will also change. When the temperature continues to rise, it is easy to fail. In the first stage of processing, it is determined based on the load average whether the generator set is continuously in a high-load state. If it is continuously in a high-load state, an early warning is issued. If it is determined in the first stage of processing that no early warning is needed, the second stage of processing is entered. In the second stage of processing, calculations are performed based on the load average and the load fluctuation vector, wherein the load fluctuation vector reflects the fluctuation of the load of the generator set over a period of time. Due to the sudden increase in load intensity, failures are also prone to occur. Therefore, this scheme further makes judgments through the second stage of processing. This scheme can further mine and calculate the sensor data to improve the timeliness of the early warning.

[0013] In some embodiments of the present invention, in the step of determining the load intensity value corresponding to each thermal imaging image, the load intensity value corresponding to each thermal imaging image is calculated based on a pre-trained first convolutional neural network model.

[0014] In some embodiments of the present invention, in the step of calculating the load average based on the load intensity values ​​corresponding to multiple consecutive time points before the current time point, and determining whether to issue a warning based on the load average, the average value of the load intensity values ​​corresponding to multiple consecutive time points before the current time point is calculated as the load average, and the load average is compared with a preset load threshold to determine whether to issue a warning.

[0015] Using the above scheme, in the first stage of processing of this scheme, the load average over a period of time is used for judgment. When the load average is large, an early warning is issued to indicate a dangerous state of the generator set, in which the generator set is prone to failure.

[0016] In some embodiments of the present invention, the step of constructing a load fluctuation vector based on load intensity values ​​corresponding to a plurality of consecutive time points before the current time point includes:

[0017] A coordinate system is constructed with time as the horizontal axis and load intensity as the vertical axis, and the position of the load intensity value corresponding to each time point is marked in the coordinate system;

[0018] Use straight lines to sequentially connect the positions of the load intensity values ​​corresponding to each time point to obtain a line graph;

[0019] A load fluctuation vector is constructed based on the line graph.

[0020] In some embodiments of the present invention, in the step of constructing a load fluctuation vector based on the line graph, the slope of the straight line between the positions of the load intensity values ​​corresponding to two adjacent time points is calculated, and the slope values ​​are used as the values ​​of one dimension of the load fluctuation vector to construct the load fluctuation vector.

[0021] Adopting the above scheme, in the second stage of processing, this scheme constructs a line graph based on the load intensity values ​​at multiple time points. The slope of the straight line between every two adjacent time points in the line graph reflects the fluctuation of the generator set at the two time points. This scheme constructs all the slope values ​​into a load fluctuation vector, which reflects the continuous fluctuation of the generator set over a period of time, thereby ensuring the accuracy of subsequent calculations.

[0022] In some embodiments of the present invention, in the step of calculating the load fluctuation value based on the load fluctuation vector, the load fluctuation vector is input into a pre-trained second convolutional neural network model, and the second convolutional neural network model outputs the load fluctuation value.

[0023] In some embodiments of the present invention, in the steps of calculating the load loss value based on the load fluctuation value and the load average, and determining whether to issue a warning based on the load loss value, a weighted calculation is performed based on the load fluctuation value and the load average to obtain the load loss value, and the load loss value is compared with a preset load health threshold to determine whether to issue a warning.

[0024] In some embodiments of the present invention, in the step of performing weighted calculation based on the load fluctuation value and the load average to obtain the load loss value, the load loss value is calculated based on the following formula:

[0025]

[0026] in, represents the load loss value, α represents the load fluctuation value, represents the load mean, a1 and a2 represent the weight values ​​corresponding to the load fluctuation value and the load mean, respectively.

[0027] Adopting the above scheme, in the second stage of calculation, on the one hand, the load fluctuation of the generator set over a period of time is calculated to determine the load fluctuation value, and on the other hand, the load fluctuation value is jointly calculated with the load average to jointly calculate the health of the generator set, thereby improving the accuracy of determining the health of the generator set, and ensuring timely early warning of the generator set when the health of the generator set is low.

[0028] Another aspect of the present invention relates to an electronic device having a computer program stored thereon, which is used to implement the steps of the aforementioned power plant early warning method based on historical data when the computer program is executed by a processor.

[0029] On the other hand, the present invention also relates to a power plant early warning system based on historical data, the system includes a computer device, the computer device includes a processor and a memory, the memory stores computer instructions, the processor is used to execute the computer instructions stored in the memory, when the computer instructions are executed by the processor, the system implements the steps implemented by the method.

[0030] In summary, the present invention has the following beneficial effects:

[0031] 1. This scheme adopts a two-stage processing method. In the first stage of processing, it is determined whether the generator set is continuously in a high-load state based on the load mean value. If it is continuously in a high-load state, an early warning is issued; if it is determined in the first stage of processing that an early warning is not required, the second stage of processing is entered. In the second stage of processing, calculation is performed based on the load mean value and the load fluctuation vector to determine whether an early warning is issued. Further mining and calculation of sensor data can be performed to improve the timeliness of early warning;

[0032] 2. In the first stage of the processing of this scheme, the load average over a period of time is used for judgment. When the load average is large, an early warning is issued to indicate a dangerous state of the generator set, in which the generator set is prone to failure;

[0033] 3. In the second stage of processing, this scheme constructs a line graph based on the load intensity values ​​at multiple time points. The slope of the straight line between every two adjacent time points in the line graph reflects the fluctuation of the generator set at the two time points. This scheme constructs all the slope values ​​into a load fluctuation vector, which reflects the continuous fluctuation of the generator set over a period of time, ensuring the accuracy of subsequent calculations;

[0034] 4. In the second stage of calculation, this solution calculates the load fluctuation of the generator set over a period of time to determine the load fluctuation value. On the other hand, the load fluctuation value is calculated together with the load average to jointly calculate the health of the generator set, thereby improving the accuracy of determining the health of the generator set and ensuring timely early warning of the generator set when the health of the generator set is low. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be 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.

[0036] Figure 1 A schematic diagram of an implementation of a power plant early warning method based on historical data of the present invention;

[0037] Figure 2 It is a schematic diagram of another implementation of the power station early warning method based on historical data of the present invention. DETAILED DESCRIPTION

[0038] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0039] The terms used in the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The singular forms "a", "the" and "the" used in the present invention and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0040] like Figure 1 As shown, the present invention provides a power plant early warning method based on historical data, the method comprising the steps of:

[0041] Step S100, obtaining thermal images taken of the generator set at each time point, and determining a corresponding load intensity value for each thermal image;

[0042] In the specific implementation process, a thermal imaging camera is pre-installed in the power plant, and the generator set is photographed by the thermal imaging camera; specifically, the generator set is photographed at each time point;

[0043] In a specific implementation process, the time interval between two adjacent time points may be 1 second, 0.5 minutes, 1 minute, or 5 minutes.

[0044] In the specific implementation process, in the step of determining the corresponding load intensity value for each thermal imaging image, the corresponding load intensity value is output through a preset convolutional neural network model.

[0045] In the specific implementation process, thermal imaging, also known as infrared thermal imaging (Far East Map), refers to the invisible light waves emitted by an object, which are converted into visible light signals after being received by a detector and processed by an electronic computer to obtain a graphical display of the temperature distribution on the surface of the object.

[0046] Thermal imaging technology is based on the fact that all objects generate heat. Infrared rays are a type of electromagnetic wave with a longer wavelength than visible light and cannot be seen directly by the human eye. However, when infrared rays hit an object, the object absorbs and reflects infrared radiation, forming hot spots or cold spots on the infrared detector. By converting these hot spots or cold spots into visible light signals and processing them with an electronic computer, a thermal image of the surface temperature distribution of the object can be obtained.

[0047] Step S200, calculating a load average based on load intensity values ​​corresponding to a plurality of consecutive time points before the current time point, and determining whether to issue an early warning based on the load average;

[0048] In some embodiments of the present invention, in the step of calculating the load average based on the load intensity values ​​corresponding to a plurality of consecutive time points before the current time point, the plurality of consecutive time points may be 10, 100 or 500 consecutive time points.

[0049] During the specific implementation process, in the step of determining whether to issue a warning based on the load average, the load average is compared with a preset load threshold to determine whether to issue a warning. Specifically, if the load average is greater than the preset load threshold, a warning is issued; if the load average is not greater than the preset load threshold, no warning is issued.

[0050] During the specific implementation process, at each time point, the load average is calculated and a determination is made whether to issue an early warning.

[0051] Step S300: If it is determined that no warning is issued based on the load average, the load average is recorded, and a load fluctuation vector is constructed based on the load intensity values ​​corresponding to a plurality of consecutive time points before the current time point, and a load fluctuation value is calculated based on the load fluctuation vector;

[0052] In the specific implementation process, the load average recorded at each time point is the average value of the load intensity values ​​at multiple consecutive time points before the time point.

[0053] Step S400, calculating a load loss value based on the load fluctuation value and the load average value, and determining whether to issue an early warning based on the load loss value.

[0054] In the specific implementation process, in the step of determining whether to issue an early warning based on the load loss value, the load loss value is compared with a preset load health threshold to determine whether to issue an early warning. If the load loss value is greater than the preset load health threshold, it is determined to issue an early warning.

[0055] In the specific implementation process, the load loss value is calculated at each time point, and whether to issue an early warning is determined based on the load loss value.

[0056] The above scheme is adopted. This scheme is a two-stage processing method. As the load of the generator set increases, the temperature of each position of the generator set will also change. When the temperature continues to rise, it is easy to fail. In the first stage of processing, it is determined based on the load average whether the generator set is continuously in a high-load state. If it is continuously in a high-load state, an early warning is issued. If it is determined in the first stage of processing that no early warning is needed, the second stage of processing is entered. In the second stage of processing, calculations are performed based on the load average and the load fluctuation vector, wherein the load fluctuation vector reflects the fluctuation of the load of the generator set over a period of time. Due to the sudden increase in load intensity, failures are also prone to occur. Therefore, this scheme further makes judgments through the second stage of processing. This scheme can further mine and calculate the sensor data to improve the timeliness of the early warning.

[0057] In some embodiments of the present invention, in the step of determining the load intensity value corresponding to each thermal imaging image, the load intensity value corresponding to each thermal imaging image is calculated based on a pre-trained first convolutional neural network model.

[0058] In the specific implementation process, the first convolutional neural network model and the second convolutional neural network model are both convolutional neural network models that have been pre-trained using training data.

[0059] In some embodiments of the present invention, in the step of calculating the load intensity value corresponding to each thermal imaging image based on a pre-trained first convolutional neural network model, the thermal imaging image can be directly input into the first convolutional neural network model, or the pixel value of each pixel point in the thermal imaging image can be extracted, and the pixel values ​​of multiple pixels are constructed as an input vector, and the input vector is input into the first convolutional neural network model.

[0060] In some embodiments of the present invention, in the step of calculating the load average based on the load intensity values ​​corresponding to multiple consecutive time points before the current time point, and determining whether to issue a warning based on the load average, the average value of the load intensity values ​​corresponding to multiple consecutive time points before the current time point is calculated as the load average, and the load average is compared with a preset load threshold to determine whether to issue a warning.

[0061] Using the above scheme, in the first stage of processing of this scheme, the load average over a period of time is used for judgment. When the load average is large, an early warning is issued to indicate a dangerous state of the generator set, in which the generator set is prone to failure.

[0062] like Figure 2 As shown, in some embodiments of the present invention, the step of constructing a load fluctuation vector based on load intensity values ​​corresponding to a plurality of consecutive time points before the current time point includes:

[0063] Step S311, constructing a coordinate system with time as the horizontal axis and load intensity as the vertical axis, and calibrating the position of the load intensity value corresponding to each time point in the coordinate system;

[0064] Step S312, using straight lines to sequentially connect the positions of the load intensity values ​​corresponding to each time point to obtain a line graph;

[0065] Step S313: construct a load fluctuation vector based on the line graph.

[0066] In a specific implementation process, in the line graph, adjacent points are connected by straight lines, and the slope of the straight line can ensure the state fluctuation of the generator set between two time points.

[0067] In some embodiments of the present invention, in the step of constructing a load fluctuation vector based on the line graph, the slope of the straight line between the positions of the load intensity values ​​corresponding to two adjacent time points is calculated, and the slope values ​​are used as the values ​​of one dimension of the load fluctuation vector to construct the load fluctuation vector.

[0068] In the specific implementation process, the formula for the slope (m) is:

[0069] Slope (m) = (y2-y1) / (x2-x1)

[0070] Among them, (x1, y1) and (x2, y2) are the coordinates of any two different points on the straight line.

[0071] Here are the steps to calculate the slope:

[0072] Determine the coordinates of the points: If the coordinates of the two points are (x1, y1) and (x2, y2).

[0073] Calculate the difference in ordinates: Calculate the difference between the ordinates of two points, that is, y2-y1.

[0074] Calculate the difference in horizontal coordinates: Calculate the difference in the horizontal coordinates of two points, that is, x2-x1.

[0075] Calculate the slope: Divide the difference in ordinates by the difference in abscissas, that is, (y2-y1) / (x2-x1). The result is the slope of the line.

[0076] Adopting the above scheme, in the second stage of processing, this scheme constructs a line graph based on the load intensity values ​​at multiple time points. The slope of the straight line between every two adjacent time points in the line graph reflects the fluctuation of the generator set at the two time points. This scheme constructs all the slope values ​​into a load fluctuation vector, which reflects the continuous fluctuation of the generator set over a period of time, thereby ensuring the accuracy of subsequent calculations.

[0077] In some embodiments of the present invention, the step of calculating the load fluctuation value based on the load fluctuation vector includes step S314, inputting the load fluctuation vector into a pre-trained second convolutional neural network model, and the second convolutional neural network model outputs the load fluctuation value.

[0078] In some embodiments of the present invention, in the steps of calculating the load loss value based on the load fluctuation value and the load average, and determining whether to issue a warning based on the load loss value, a weighted calculation is performed based on the load fluctuation value and the load average to obtain the load loss value, and the load loss value is compared with a preset load health threshold to determine whether to issue a warning.

[0079] During the specific implementation process, the load fluctuation value and the load average are both provided with corresponding weight values. In the step of performing weighted calculation based on the load fluctuation value and the load average to obtain the load loss value, the load loss value is calculated based on the corresponding weight values ​​set for the load fluctuation value and the load average and the load fluctuation value and the load average. Specifically, a weighted average value can be adopted.

[0080] In some embodiments of the present invention, in the step of performing weighted calculation based on the load fluctuation value and the load average to obtain the load loss value, the load loss value is calculated based on the following formula:

[0081]

[0082] in, represents the load loss value, α represents the load fluctuation value, represents the load mean, a1 and a2 represent the weight values ​​corresponding to the load fluctuation value and the load mean, respectively.

[0083] Adopting the above scheme, in the second stage of calculation, on the one hand, the load fluctuation of the generator set over a period of time is calculated to determine the load fluctuation value, and on the other hand, the load fluctuation value is jointly calculated with the load average to jointly calculate the health of the generator set, thereby improving the accuracy of determining the health of the generator set, and ensuring timely early warning of the generator set when the health of the generator set is low.

[0084] Another aspect of the present invention relates to an electronic device having a computer program stored thereon, which, when executed by a processor, implements the steps of the power plant early warning method based on historical data. The electronic device may be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the technical field.

[0085] An embodiment of the present invention also provides a power plant early warning system based on historical data, the system comprising a computer device, the computer device comprising a processor and a memory, the memory storing computer instructions, the processor being used to execute the computer instructions stored in the memory, and when the computer instructions are executed by the processor, the system implements the steps implemented by the method.

[0086] It should be understood by those skilled in the art that the exemplary components, systems and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software or a combination of the two. Whether it is performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link via a data signal carried in a carrier.

[0087] It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present invention.

[0088] In the present invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with features of other embodiments or replace features of other embodiments.

[0089] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the embodiments of the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A power plant early warning method based on historical data, characterized in that: The steps of the method include: Obtain thermal images taken of the generator set at each time point, and determine the load intensity value corresponding to each thermal image; Calculate the load average based on the load intensity values ​​corresponding to multiple consecutive time points before the current time point, and determine whether to issue an early warning based on the load average; If it is determined that no warning is issued based on the load average, the load average is recorded, and a load fluctuation vector is constructed based on the load intensity values ​​corresponding to multiple consecutive time points before the current time point, and the load fluctuation value is calculated based on the load fluctuation vector; A load loss value is calculated based on the load fluctuation value and the load average value, and whether to issue an early warning is determined based on the load loss value.

2. The power station early warning method based on historical data according to claim 1 is characterized by: In the step of determining the load intensity value corresponding to each thermal imaging image, the load intensity value corresponding to each thermal imaging image is calculated based on a pre-trained first convolutional neural network model.

3. The power station early warning method based on historical data according to claim 1 is characterized by: In the step of calculating the load average based on the load intensity values ​​corresponding to multiple consecutive time points before the current time point, and determining whether to issue a warning based on the load average, the average value of the load intensity values ​​corresponding to multiple consecutive time points before the current time point is calculated as the load average, and the load average is compared with a preset load threshold to determine whether to issue a warning.

4. The power station early warning method based on historical data according to any one of claims 1 to 3, characterized in that: The steps of constructing a load fluctuation vector based on the load intensity values ​​corresponding to a plurality of consecutive time points before the current time point include: A coordinate system is constructed with time as the horizontal axis and load intensity as the vertical axis, and the position of the load intensity value corresponding to each time point is marked in the coordinate system; Use straight lines to sequentially connect the positions of the load intensity values ​​corresponding to each time point to obtain a line graph; A load fluctuation vector is constructed based on the line graph.

5. The power station early warning method based on historical data according to claim 4 is characterized by: In the step of constructing a load fluctuation vector based on the line graph, the slope of the straight line between the positions of the load intensity values ​​corresponding to two adjacent time points is calculated, and the slope values ​​are used as the values ​​of one dimension of the load fluctuation vector to construct the load fluctuation vector.

6. The power station early warning method based on historical data according to claim 1 is characterized by: In the step of calculating the load fluctuation value based on the load fluctuation vector, the load fluctuation vector is input into a pre-trained second convolutional neural network model, and the second convolutional neural network model outputs the load fluctuation value.

7. The power station early warning method based on historical data according to claim 1 is characterized by: In the step of calculating the load loss value based on the load fluctuation value and the load average, and determining whether to issue a warning based on the load loss value, a weighted calculation is performed based on the load fluctuation value and the load average to obtain the load loss value, and the load loss value is compared with a preset load health threshold to determine whether to issue a warning.

8. The power station early warning method based on historical data according to claim 7 is characterized by: In the step of performing weighted calculation based on the load fluctuation value and the load average to obtain the load loss value, the load loss value is calculated based on the following formula: in, represents the load loss value, α represents the load fluctuation value, represents the load mean, a1 and a2 represent the weight values ​​corresponding to the load fluctuation value and the load mean, respectively.

9. An electronic device, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the steps of the power plant early warning method based on historical data as described in any one of claims 1 to 8 are implemented.

10. A power plant early warning system based on historical data, characterized in that: The system includes a computer device, which includes a processor and a memory. The memory stores computer instructions. The processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps implemented by the power plant early warning method based on historical data as described in any one of claims 1 to 8.

Images