Distribution network operation state prediction method and system

Abstract

The invention discloses a distribution network operation state prediction method and system, and relates to the technical field of intelligent prediction, and the method comprises the following steps: constructing an open-loop equation, inputting interference variables and input variables into the open-loop equation to obtain corresponding output variables, carrying out the analysis to obtain the periodic characteristics and probability characteristics of each interference variable, and carrying out the prediction of the operation state of a distribution network. And outputting the interference variable and the corresponding probability. By analyzing the period and probability characteristics in the historical data, the influence of the interference variables on the running state of the distribution network can be more accurately predicted, multiple interference variables and dynamic changes thereof are considered, uncertainty and complex working conditions can be better dealt with, the occurrence probability of the interference variables is predicted in advance, maintenance and resource allocation are reasonably arranged, and the reliability of the distribution network is improved. The operation efficiency is improved, data support is provided for distribution network operation management personnel, the distribution network operation management personnel are helped to make more scientific decisions, feature analysis can be updated in real time through the sliding window technology, and dynamic changes of the distribution network operation state can be adapted.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent prediction, and in particular to a distribution network operation status prediction method and system. Background Art

[0002] In recent years, data technology has become an important support for the prediction of distribution network operation status. By collecting and analyzing massive amounts of operation data, including voltage, current, power, temperature and other information, load prediction, fault prediction and energy management functions can be realized. The application of big data technology not only improves the accuracy and practicality of prediction, but also provides a scientific basis for the operation and maintenance of power systems. Driven by emerging technologies such as big data, artificial intelligence, and digital twins, distribution network operation status prediction technology has made significant progress, but it still needs to be continuously optimized and innovated to meet the development needs of future power systems.

[0003] At present, in the Chinese invention patent with publication number CN117277266A, a distribution network operation status prediction method based on the improved firefly algorithm to optimize the RBF neural network is disclosed. The method constructs a distribution network operation status prediction model; inputs the divided training set into the distribution network operation status prediction model for training, and uses the mean absolute error and the root mean square error as prediction evaluation indicators; compares the voltage prediction value with the actual voltage change of each node. If the voltage difference is less than the threshold, the distribution network operation status is considered normal; if the voltage difference is greater than the threshold, the distribution network operation status is unstable. However, the relevant technology does not predict the relevant data of the power grid based on historical data, lacks the scientific nature of the prediction, and does not optimize the output of the prediction result based on the probability and frequency of the interference, which is easy to lead to misleading predictions and is not conducive to the inclusiveness and accuracy of the prediction. Summary of the invention

[0004] The technical problem solved by the present invention is that the relevant data of the power grid is not predicted based on historical data in the related technology, the prediction lacks scientificity, and the prediction results are not optimized and output based on the probability and frequency of occurrence of interference, which can easily lead to misleading predictions and is not conducive to the inclusiveness and accuracy of the prediction.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: In the first aspect, a distribution network operation status prediction method comprises the following steps: Step S100, obtaining input quantity, output quantity and interference quantity according to historical operation data, constructing an open-loop equation according to the input quantity, output quantity and interference quantity, and obtaining the corresponding output quantity by inputting the interference quantity and input quantity into the open-loop equation; Step S200, setting a time window, sliding the interference amount according to the time window, obtaining interference sub-data segments, analyzing the sub-data segments, and obtaining periodic characteristics and probability characteristics of each interference amount; Step S300, obtaining the current time point, matching the corresponding interference amount according to the periodic characteristics of the interference amount, matching the probability of occurrence of the interference amount according to the probability characteristics, and outputting the interference amount and the corresponding probability.

[0006] As a preferred solution of the distribution network operation status prediction method described in the present invention, the historical operation data includes input power, ambient temperature, ambient humidity, ambient wind speed, output power, cable aging degree and transformer power conversion efficiency; The reference value of the response speed of the cable transmission power is a first value, and the calculation method of the cable aging degree includes: Calculate the difference between the response speed of the cable transmission power and the first value, record it as the first difference, calculate the ratio of the first difference to the first value, and set the ratio of the first difference to the first value as the cable aging degree; The transformer power conversion efficiency is expressed as the ratio of the transformer output power to the transformer input power.

[0007] As a preferred scheme of a distribution network operation status prediction method described in the present invention, the input power is set as the input quantity, the ambient temperature, ambient humidity, ambient wind speed, cable aging degree and transformer power conversion efficiency are set as interference quantities, and the output power is set as the output quantity. The input quantity, output quantity and interference quantity are time domain signals and change according to the change of time.

[0008] As a preferred solution of the distribution network operation state prediction method described in the present invention, the calculation method of the open-loop equation includes: Through discrete Fourier transform, the input, output and interference are converted from time domain signals to frequency domain signals, the interference is ignored, an open-loop transfer function is constructed based on the converted input and output, the interference is added, and the open-loop transfer function is converted into an open-loop equation.

[0009] As a preferred solution of the distribution network operation state prediction method described in the present invention, the calculation expression of the open-loop equation is: ; in, is the output, is the input quantity, , , , and are interference quantities, and are respectively ambient temperature, ambient humidity, ambient wind speed, cable aging degree and transformer power conversion efficiency. is the open-loop transfer function, , , , and is the influence coefficient corresponding to each interference quantity, and the influence coefficient is obtained through experimental calculation.

[0010] As a preferred solution of the distribution network operation status prediction method described in the present invention, wherein: the first time length is set to the time window length, and the second value is set to the time step length; Arrange the interference quantity into a time series format, and use a sliding window technique to continuously move the window from the starting point of the time series, each time moving the length of the second value, extract data fragments from each window, and set the data fragments as interference quantum data fragments; The sub-data segments are analyzed to obtain the periodic characteristics and probability characteristics of each interference quantity.

[0011] As a preferred solution of a distribution network operation status prediction method described in the present invention, wherein: a discrete Fourier transform is performed on each sub-data segment, time domain data is converted into frequency domain data, the power spectrum density of the transformed interference amount is calculated, a third value is set as a reference value of the power spectrum density, each power spectrum density is compared with the third value, a power spectrum density segment that is greater than or equal to the third value is selected, the time interval between the end point of the previous segment and the initial point of the next segment of adjacent power spectrum density segments is calculated, each time interval is traversed, the average value of the time interval is calculated, the average value of the time interval is set as the periodic characteristic of the corresponding interference amount, and a first mapping relationship between the interference amount and the periodic characteristic is established.

[0012] As a preferred scheme of a distribution network operation status prediction method described in the present invention, the total number of sub-data segments is counted, the number of times each interference quantity appears in the sub-data segment is counted, recorded as the first number, the ratio of the first number to the total number of sub-data segments is calculated, and it is set as the probability feature of the corresponding interference quantity, and a second mapping relationship between the interference quantity and the probability feature is established.

[0013] As a preferred solution of a distribution network operation state prediction method described in the present invention, wherein: the current time point is obtained, the time interval from the current time point to the end point of the previous sub-data segment is calculated, recorded as the first interval, the difference between the first interval and the periodic characteristics corresponding to each interference amount is calculated, the difference between the first interval with the smallest absolute value and the periodic characteristics corresponding to each interference amount is selected, recorded as the target difference, and the interference amount corresponding to the target difference is set as the predicted interference amount; Inputting the predicted interference amount into the second mapping relationship to match the corresponding probability; The interference amount and the corresponding probability are output.

[0014] In a second aspect, a distribution network operation status prediction system includes an analysis module, a calculation module and a mapping module; The analysis module is used to obtain input quantity, output quantity and interference quantity according to historical operation data, construct an open-loop equation according to the input quantity, output quantity and interference quantity, and obtain the corresponding output quantity by inputting the interference quantity and input quantity into the open-loop equation; The calculation module is used to set a time window, slide the interference amount according to the time window, obtain interference quantum data fragments, analyze the sub-data fragments, and obtain the periodic characteristics and probability characteristics of each interference amount; The mapping module is used to obtain the current time point, match the corresponding interference amount according to the periodic characteristics of the interference amount, match the probability of occurrence of the interference amount according to the probability characteristics, and output the interference amount and the corresponding probability.

[0015] The beneficial effects of the present invention are as follows: by analyzing the periodic and probabilistic characteristics in historical data, the impact of interference on the operating status of the distribution network can be predicted more accurately, and multiple interference quantities and their dynamic changes are taken into account, so that uncertainty and complex working conditions can be better dealt with, the probability of occurrence of interference quantities can be predicted in advance, maintenance and resource allocation can be reasonably arranged, and operating efficiency can be improved. Data support is provided for distribution network operation managers to help them make more scientific decisions. Through the sliding window technology, feature analysis can be updated in real time to adapt to the dynamic changes in the operating status of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of the basic flow of a distribution network operation status prediction method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0018] Example, see Figure 1 , as an embodiment of the present invention, provides a distribution network operation status prediction method, comprising the following steps: Step S100, obtaining input quantity, output quantity and interference quantity according to historical operation data, constructing an open-loop equation according to the input quantity, output quantity and interference quantity, and obtaining the corresponding output quantity by inputting the interference quantity and input quantity into the open-loop equation; Step S200, setting a time window, sliding the interference amount according to the time window, obtaining interference sub-data segments, analyzing the sub-data segments, and obtaining periodic characteristics and probability characteristics of each interference amount; Step S300, obtaining the current time point, matching the corresponding interference amount according to the periodic characteristics of the interference amount, matching the probability of occurrence of the interference amount according to the probability characteristics, and outputting the interference amount and the corresponding probability.

[0019] By analyzing the periodic and probability characteristics in historical data, the present invention can more accurately predict the impact of interference on the operating status of the distribution network, taking into account a variety of interference quantities and their dynamic changes, and can better cope with uncertainty and complex working conditions, predict the probability of occurrence of interference quantities in advance, reasonably arrange maintenance and resource allocation, improve operating efficiency, provide data support for distribution network operation managers, and help them make more scientific decisions. Through the sliding window technology, the feature analysis can be updated in real time to adapt to the dynamic changes in the operating status of the distribution network.

[0020] The historical operation data includes input power, ambient temperature, ambient humidity, ambient wind speed, output power, cable aging and transformer power conversion efficiency; The reference value of the response speed of the cable transmission power is a first value, and the calculation method of the cable aging degree includes: Calculate the difference between the response speed of the cable transmission power and the first value, record it as the first difference, calculate the ratio of the first difference to the first value, and set the ratio of the first difference to the first value as the cable aging degree; The transformer power conversion efficiency is expressed as the ratio of the transformer output power to the transformer input power.

[0021] In the specific implementation, by analyzing the periodic and probability characteristics in the historical data, it is possible to more accurately predict the impact of interference on the operating status of the distribution network. Through dynamic monitoring of cable aging and transformer efficiency, potential problems can be discovered in advance and sudden failures can be reduced. The historical operating data are as follows: input power: 100kW, output power: 95 kW, ambient temperature: 30°C, ambient humidity: 70%, ambient wind speed: 5m / s, cable aging: 0.1, transformer efficiency: 95%, cable aging: assuming the response speed is 0.9 and the reference value is 1.0, the aging degree is 0.1, indicating that the cable aging degree is relatively light, and the transformer efficiency is 95%, indicating that the transformer is in good operating condition.

[0022] The input power is set as the input quantity, the ambient temperature, ambient humidity, ambient wind speed, cable aging degree and transformer power conversion efficiency are set as the interference quantity, and the output power is set as the output quantity. The input quantity, output quantity and interference quantity are time domain signals and change according to the change of time.

[0023] In specific implementation, through frequency domain signal analysis, the dynamic changes of input and interference can be captured, the accuracy of output power prediction can be improved, and the extraction of periodic characteristics and probability characteristics enables the model to better adapt to complex operating environments. The sliding window technology allows the model to be updated in real time to adapt to the dynamic changes in the distribution network operating status. By predicting the changes in output power, potential problems can be discovered in advance, and maintenance and resource allocation can be reasonably arranged.

[0024] The calculation method of the open-loop equation includes: Through discrete Fourier transform, the input, output and interference are converted from time domain signals to frequency domain signals, the interference is ignored, an open-loop transfer function is constructed based on the converted input and output, the interference is added, and the open-loop transfer function is converted into an open-loop equation.

[0025] The calculation expression of the open-loop equation is: ; in, is the output, is the input quantity, , , , and are interference quantities, and are respectively ambient temperature, ambient humidity, ambient wind speed, cable aging degree and transformer power conversion efficiency. is the open-loop transfer function, , , , and is the influence coefficient corresponding to each interference quantity, and the influence coefficient is obtained through experimental calculation.

[0026] In specific implementation, frequency domain analysis can more effectively capture the impact of input and interference on output, thereby improving prediction accuracy. The introduction of open-loop transfer function and interference transfer function enables the system to analyze stability more intuitively in the frequency domain. Frequency domain analysis provides a theoretical basis for the design and optimization of control systems, and helps to formulate more effective control strategies. Using FFT for frequency domain conversion can significantly reduce computational complexity and improve computational efficiency.

[0027] The first time length is set as the time window length, and the second value is set as the time step length; Arrange the interference quantity into a time series format, and use a sliding window technique to continuously move the window from the starting point of the time series, each time moving the length of the second value, extract data fragments from each window, and set the data fragments as interference quantum data fragments; The sub-data segments are analyzed to obtain the periodic characteristics and probability characteristics of each interference quantity.

[0028] Perform a discrete Fourier transform on each sub-data segment, convert the time domain data into frequency domain data, calculate the power spectrum density of the transformed interference amount, set the third value as a reference value of the power spectrum density, compare each power spectrum density with the third value, select a power spectrum density segment that is greater than or equal to the third value, calculate the time interval between the end point of the previous segment and the initial point of the next segment of adjacent power spectrum density segments, traverse each time interval, calculate the average value of the time interval, set the average value of the time interval as the periodic characteristic of the corresponding interference amount, and establish a first mapping relationship between the interference amount and the periodic characteristic.

[0029] Count the total number of sub-data segments, count the number of times each interference amount appears in the sub-data segment, record it as the first number, calculate the ratio of the first number to the total number of sub-data segments, set it as the probability feature of the corresponding interference amount, and establish a second mapping relationship between the interference amount and the probability feature.

[0030] Obtain the current time point, calculate the time interval from the current time point to the end point of the previous sub-data segment, record it as the first interval, calculate the difference between the first interval and the periodic characteristics corresponding to each interference amount, select the difference between the first interval and the periodic characteristics corresponding to each interference amount with the smallest absolute value, record it as the target difference, and set the interference amount corresponding to the target difference as the predicted interference amount; Inputting the predicted interference amount into the second mapping relationship to match the corresponding probability; The interference amount and the corresponding probability are output.

[0031] In specific implementation, the extraction of periodic features helps to identify potential interference patterns in advance, thereby optimizing the system operation strategy and enhancing stability. Through statistical analysis, the probability of interference can be estimated more accurately, thereby improving the prediction accuracy of output power. The extraction of probabilistic features helps to identify potential interference patterns in advance, thereby optimizing the system operation strategy and enhancing stability.

[0032] By analyzing the periodic and probability characteristics in historical data, the present invention can more accurately predict the impact of interference on the operating status of the distribution network, taking into account a variety of interference quantities and their dynamic changes, and can better cope with uncertainty and complex working conditions, predict the probability of occurrence of interference quantities in advance, reasonably arrange maintenance and resource allocation, improve operating efficiency, provide data support for distribution network operation managers, and help them make more scientific decisions. Through the sliding window technology, the feature analysis can be updated in real time to adapt to the dynamic changes in the operating status of the distribution network.

[0033] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program codes. Wherein, the storage medium is implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, magnetic disk or optical disk. These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0034] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions of the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A distribution network operation status prediction method, characterized in that: The following steps are involved: Step S100, obtaining input quantity, output quantity and interference quantity according to historical operation data, constructing an open-loop equation according to the input quantity, output quantity and interference quantity, and obtaining the corresponding output quantity by inputting the interference quantity and input quantity into the open-loop equation; Step S200, setting a time window, sliding the interference amount according to the time window, obtaining interference sub-data segments, analyzing the sub-data segments, and obtaining periodic characteristics and probability characteristics of each interference amount; Step S300, obtaining the current time point, matching the corresponding interference amount according to the periodic characteristics of the interference amount, matching the probability of occurrence of the interference amount according to the probability characteristics, and outputting the interference amount and the corresponding probability.

2. A distribution network operation status prediction method according to claim 1, characterized in that: The historical operation data includes input power, ambient temperature, ambient humidity, ambient wind speed, output power, cable aging and transformer power conversion efficiency; The reference value of the response speed of the cable transmission power is a first value, and the calculation method of the cable aging degree includes: Calculate the difference between the response speed of the cable transmission power and the first value, record it as the first difference, calculate the ratio of the first difference to the first value, and set the ratio of the first difference to the first value as the cable aging degree; The transformer power conversion efficiency is expressed as the ratio of the transformer output power to the transformer input power.

3. A distribution network operation status prediction method according to claim 1, characterized in that: The input power is set as the input quantity, the ambient temperature, ambient humidity, ambient wind speed, cable aging degree and transformer power conversion efficiency are set as the interference quantity, and the output power is set as the output quantity. The input quantity, output quantity and interference quantity are time domain signals and change according to the change of time.

4. A distribution network operation status prediction method according to claim 1, characterized in that: The calculation method of the open-loop equation includes: Through discrete Fourier transform, the input, output and interference are converted from time domain signals to frequency domain signals, the interference is ignored, an open-loop transfer function is constructed based on the converted input and output, the interference is added, and the open-loop transfer function is converted into an open-loop equation.

5. A distribution network operation status prediction method according to claim 4, characterized in that: The calculation expression of the open-loop equation is: ； in, is the output, is the input quantity, , , , and are interference quantities, and are respectively ambient temperature, ambient humidity, ambient wind speed, cable aging degree and transformer power conversion efficiency. is the open-loop transfer function, , , , and is the influence coefficient corresponding to each interference quantity, and the influence coefficient is obtained through experimental calculation.

6. A distribution network operation status prediction method according to claim 1, characterized in that: The first time length is set as the time window length, and the second value is set as the time step length; Arrange the interference quantity into a time series format, and use a sliding window technique to continuously move the window from the starting point of the time series, each time moving the length of the second value, extract data fragments from each window, and set the data fragments as interference quantum data fragments; The sub-data segments are analyzed to obtain the periodic characteristics and probability characteristics of each interference quantity.

7. A distribution network operation status prediction method according to claim 6, characterized in that: Perform a discrete Fourier transform on each sub-data segment, convert the time domain data into frequency domain data, calculate the power spectrum density of the transformed interference amount, set the third value as a reference value of the power spectrum density, compare each power spectrum density with the third value, select a power spectrum density segment that is greater than or equal to the third value, calculate the time interval between the end point of the previous segment and the initial point of the next segment of adjacent power spectrum density segments, traverse each time interval, calculate the average value of the time interval, set the average value of the time interval as the periodic characteristic of the corresponding interference amount, and establish a first mapping relationship between the interference amount and the periodic characteristic.

8. A distribution network operation status prediction method according to claim 6, characterized in that: Count the total number of sub-data segments, count the number of times each interference amount appears in the sub-data segment, record it as the first number, calculate the ratio of the first number to the total number of sub-data segments, set it as the probability feature of the corresponding interference amount, and establish a second mapping relationship between the interference amount and the probability feature.

9. A distribution network operation status prediction method according to claim 1, characterized in that: Obtain the current time point, calculate the time interval from the current time point to the end point of the previous sub-data segment, record it as the first interval, calculate the difference between the first interval and the periodic characteristics corresponding to each interference amount, select the difference between the first interval and the periodic characteristics corresponding to each interference amount with the smallest absolute value, record it as the target difference, and set the interference amount corresponding to the target difference as the predicted interference amount; Inputting the predicted interference amount into the second mapping relationship to match the corresponding probability; The interference amount and the corresponding probability are output.

10. A distribution network operation status prediction system, the system is used to execute the distribution network operation status prediction method according to claim 1, characterized in that: It includes an analysis module, a calculation module and a mapping module; The analysis module is used to obtain input quantity, output quantity and interference quantity according to historical operation data, construct an open-loop equation according to the input quantity, output quantity and interference quantity, and obtain the corresponding output quantity by inputting the interference quantity and input quantity into the open-loop equation; The calculation module is used to set a time window, slide the interference amount according to the time window, obtain interference quantum data fragments, analyze the sub-data fragments, and obtain the periodic characteristics and probability characteristics of each interference amount; The mapping module is used to obtain the current time point, match the corresponding interference amount according to the periodic characteristics of the interference amount, match the probability of occurrence of the interference amount according to the probability characteristics, and output the interference amount and the corresponding probability.

Images