An intelligent new energy grid-connected state monitoring method and system

Abstract

The application discloses a kind of intelligent new energy grid-connected state monitoring method and system, it is related to power grid data management technical field, obtains the periodic variation law of the first target unit operating state, extracts the law feature in periodic variation law in new energy grid-connected power generation system, obtains the same periodic variation law with the first target unit operating state and is recorded as reference unit generator set, in the same variation period, the difference of the target parameter of reference unit generator set and the first generator set is used as difference evaluation value, obtain a subsystem consisting of several generator sets and an upper connection equipment of a generator set, by calculating the similarity of the operating parameter variation of each generator set and upper connection equipment, the generator set with the highest degree of association with upper connection equipment in several generator sets is set as monitored unit, obtain the reference unit of monitored unit, compare the real-time operating data of monitored unit and the reference unit of monitored unit.

Description

Technical Field

[0001] This invention relates to the field of power grid data management technology, specifically to an intelligent method and system for monitoring the grid connection status of new energy sources. Background Technology

[0002] Compared to traditional power generation methods, the single-unit power output of new energy power generation units such as photovoltaic and wind power is limited. Therefore, multiple power generation units are typically networked to increase the overall power generation capacity. Because multiple redundant power generation units exist within a new energy power generation system, different units share the same electrical data parameters. During real-time monitoring of the system's operational data, personnel may lack awareness of faulty operational data, overlooking or neglecting abnormal data from certain units. Consequently, when abnormal data is not readily apparent within the system, it is difficult to detect, posing a potential threat to the system's operation. Summary of the Invention

[0003] The purpose of this invention is to provide an intelligent method and system for monitoring the grid connection status of new energy sources, so as to solve the problems raised in the prior art.

[0004] To achieve the above objectives, the present invention provides an intelligent method for monitoring the grid connection status of new energy sources, as follows: Step S100: In a new energy grid-connected power generation system including several generator sets, a certain generator set is set as the first target unit, and a certain operating parameter of the first target unit is used as the target parameter. From the historical records of the target parameter, the periodic change pattern of the operating status of the first target unit is obtained, and the regular characteristics in the periodic change pattern are extracted. Step S200: In the new energy grid-connected power generation system, the generator set that has the same periodic change pattern as the first target unit is recorded as the reference unit. In the operation records of the first target unit and the reference unit, the time interval in which the same pattern characteristics appear is obtained. In the same change cycle, the difference between the target parameters of the reference unit and the first generator set is used as the difference evaluation value. Step S300: In the new energy grid-connected power generation system, a subsystem consisting of several generator sets and an upstream device of the generator set is obtained. Real-time operation records in the subsystem are collected. By calculating the similarity of the changes in the operating parameters of each generator set and the upstream device, the generator set with the highest correlation with the upstream device among the several generator sets is set as the monitored generator set. Step S400: Obtain the reference unit of the monitored unit, compare the real-time operating data of the monitored unit with the reference unit of the monitored unit, and when the deviation between the difference evaluation value corresponding to the real-time operating data and the difference evaluation value corresponding to the real-time operating data exceeds the threshold, issue a prompt to the relevant management personnel.

[0005] Furthermore, step S100 includes: Step S101: Obtain the operation record of the target parameters of the first target unit within the time period T, set the unit sampling period to T0, and collect the target parameters once after each unit sampling period to obtain one sample data. Step S102: Obtain N consecutive sampling data points, arrange the N sampling data points according to the sampling time sequence to obtain the sampling sequence L, and form the second sampling period by N2 sampling positions, where the time interval between any two adjacent sampling positions is T0, and N>N2; Step S103: The first sampling position of the second sampling period is associated with the nk-th sampled data in the sampling sequence L. Each sampling position in the second sampling period is associated with the sampled data in the sampling sequence in turn. The sampled data at the corresponding position in the sampling sequence L is obtained at each sampling position and is denoted as the first sampling sequence of the second sampling period. The average value of the sampled data in the first sampling sequence of the second sampling period is calculated and denoted as A21. Step S104: Record the first nk-1 sampled data in the sampling sequence L as the first sampling sequence of the first sampling period, calculate the average value of the sampled data in the first sampling sequence of the first sampling period, and record it as A11. Move the second sampling sequence by 1 unit sampling period along the time order of the sampled data in the sampling sequence, and repeat the method in step S103 to obtain the second sampling sequence of the second sampling period. Calculate the average value of the sampled data in the second sampling sequence of the second sampling period and record it as A22. Record the first nk sampled data in the sampling sequence as the second sampling sequence of the first sampling period, calculate the average value of the sampled data in the second sampling sequence of the first sampling period, and record it as A12. Step S105: Match the m-th sampling sequence of the first sampling period with the m-th sampling sequence of the second sampling period, and calculate the difference Gm between the average values ​​of the sampled data, Gm=|A1m-A2m|, where A1m represents the average value of the sampled data in the m-th sampling sequence of the first sampling period, and A2m represents the average value of the sampled data in the m-th sampling sequence of the second sampling period; Step S106: Collect the difference between the average values ​​of c sampled data, obtain the largest difference from the differences between the average values ​​of c sampled data, denoted as Gmo, and obtain the position of the mo-th sampled sequence in the sampling sequence of the second sampling period, where the first sampled data in the mo-th sampled sequence of the second sampling period corresponds to the ko-th sampled data in the sampling sequence L. Step S107: In the sampling sequence L, the total sampling time corresponding to the first ko-1 sampled data is collected to obtain the target sampling period. The time length of the target sampling period of the first target unit is recorded as T1. The unit sampling period between the ko-1 sampled data and the ko sampled data in the sampling sequence L is obtained, and the midpoint of the unit sampling period is recorded as the transition time. New energy generator sets, such as photovoltaic power generation and wind power generation, have strong time periodicity, which is divided into peak power generation period and low power generation period. By calculating the average power generation in different time periods, the power generation cycle variation pattern of generator sets can be captured, and the dividing point between peak power generation period and low power generation period can be recorded as the transition time. By obtaining a generator set that has the same variation pattern as the first target generator set as the reference generator set, the variation time of the peak generation period and the trough generation period of the reference generator set is consistent with the variation time of the first target generator set.

[0006] Furthermore, step S200 includes: Step S201: In the new energy grid-connected power generation system, a generator set that is different from the first target generator set is designated as the second target generator set. The operation record of the target parameters of the second target generator set within the time period T is obtained. The steps in step S100 are repeated to obtain the target sampling period of the second target generator set. The time length of the target sampling period of the second target generator set is denoted as T2. When T1=T2, the second target generator set is the reference generator set of the first target generator set. The corresponding transition time of the reference generator set is obtained. Step S202: Calculate the reference delay , Where t1 represents the transition time corresponding to the first target unit, and t2 represents the transition time corresponding to the second target unit; In order to obtain more reference units for the first target generator set, the requirements for the reference units include that the change pattern of the target parameters is the same as the periodic change pattern of the first target generator set. In order to better obtain the reference units and reduce the difference in the time sequence of the same periodic change pattern, as long as the same change pattern can be met, they are adopted. The reference delay is used to align the same periodic change patterns. Step S203: Plot the function graph of the target parameters changing with time for the first target unit in the corresponding target sampling period to obtain the first parameter image; plot the function graph of the target parameters changing with time for the second target unit in the corresponding target sampling period to obtain the second parameter image; plot the first parameter image and the second parameter image in the same coordinate system. Step S204: In the same coordinate system, obtain the point corresponding to any sampled data in the second parameter image, obtain the shortest distance from the point to the first parameter image, sum the shortest distances from all sampled data points to the first parameter image, and obtain the difference evaluation value between the reference unit and the first target unit. The graphs of two periodically changing functions are periodically aligned. The difference between the sampled data points in the second reference image and the first reference image is taken as the difference between the sampled data. The shortest distances between each sampled data point in the second parameter image and the first reference image are accumulated to obtain the evaluation value of the difference between the second reference image and the first parameter image.

[0007] Furthermore, step S300 includes: Step S301: Obtain the real-time operating data of the target parameter of a certain generator set in the subsystem. Within a time period of TC, obtain the real-time operating data, convert the real-time operating data into a data distribution column, denoted as data distribution column P1, and obtain the measurement range of the target parameter in a certain generator set, denoted as the first measurement interval. Step S302: Obtain the upstream device connected to a certain generator set, obtain the parameter item corresponding to the target parameter in the upstream device, record the parameter item as the second target parameter, perform linear transformation or nonlinear transformation on the second target parameter, represent the real-time running data of the second target parameter in TC in the first measurement interval, convert the running data into a data distribution column, and obtain the data distribution column Q1, where the start and end times of the time period with a time length of TC are the same. Step S303: Take distribution column Q1 as the reference distribution and distribution column P1 as the fitting distribution. Fit the reference distribution with the fitting distribution and calculate the information loss generated during the fitting process. The information loss is the difference in information entropy between the fitting distribution and the reference distribution. Step S304: Traverse the real-time operating data of all generator sets in the subsystem, obtain the information loss between the real-time operating data of each generator set and the real-time operating data of the second target parameter of the upstream equipment, and set the generator set with the minimum information loss as the monitored unit. In one subsystem, the second target parameter of the uplink device is affected by the operating status of the target parameters of several generator sets. The operating data of the target parameters of the generator sets are used as a code. The code formed by the operating data of the second target parameter is fitted to obtain the number of bits added to each symbol in the code corresponding to the operating data of the target parameter. The number of bits added to each symbol in the operating data of the target parameters of the two types of generator sets is compared. The distribution with the smaller number of bits added is closer to the code formed by the operating data of the second target parameter. The number of bits added is the difference in information entropy from P1 to Q1.

[0008] Furthermore, step S400 includes: Step S401: Obtain the target sampling period corresponding to the controlled unit and the duration of the target sampling period; obtain the reference delay between the monitored unit and its reference unit, denoted as... The difference evaluation value d0 between the monitored unit and the reference unit of the monitored unit; Step S402: From At the start of the event, real-time operation records of the target parameters of the monitored unit are collected. The collection duration is the length of the target sampling period. A graph of the real-time operation records of the target parameters as a function of time is plotted and recorded as the first operation graph. Step S403: From Starting at a certain time, real-time operational records of the target parameters of the monitored unit and the reference unit are collected. The collection duration is the length of the target sampling period. A graph of the real-time operational records of the target parameters as a function of time is plotted, denoted as the second operational graph. ; Step S404: After each unit sampling period, take a real-time sampling point on the second running image, calculate the shortest distance from each real-time sampling point to the first running image, and sum the shortest distances to obtain the current distance. ,when When this happens, a prompt is made to the generator set management personnel, where H represents the difference judgment value. , Indicates obtaining The maximum value in d0.

[0009] To better implement the above methods, an intelligent new energy grid-connected status monitoring system is proposed. The system includes: a change pattern management module, a difference evaluation value management module, a real-time monitoring module, and a comparison management module. The change pattern management module is used to collect historical data to manage the periodic change patterns of generator sets. The difference evaluation value management module is used to compare the operating parameters of the reference unit and the first target unit and calculate the difference in target parameters. The real-time monitoring module is used to manage the real-time operating status of generator sets. The comparison management module is used to compare the real-time operating data of the monitored unit with the reference unit of the monitored unit.

[0010] Furthermore, the change pattern management module includes: a data acquisition unit, a sampling period management unit, a sampling movement unit, a difference comparison unit, and a transition time extraction unit. The data acquisition unit is used to collect the operating data of the generator set, the sampling period management unit is used to obtain the second sampling period, the sampling movement unit is used to move the second sampling period in the sampling sequence, the difference comparison unit is used to calculate the difference between the average values ​​of the sampled data, and the transition time extraction unit is used to obtain the transition time.

[0011] Furthermore, the change pattern management module includes: a data acquisition unit, a sampling period management unit, a sampling movement unit, a difference comparison unit, and a transition time extraction unit. The data acquisition unit is used to collect the operating data of the generator set, the sampling period management unit is used to obtain the second sampling period, the sampling movement unit is used to move the second sampling period in the sampling sequence, the difference comparison unit is used to calculate the difference between the average values ​​of the sampled data, and the transition time extraction unit is used to obtain the transition time.

[0012] Furthermore, the real-time monitoring module includes: a subsystem monitoring unit, a distribution acquisition unit, a distribution fitting unit, and a difference comparison unit. The subsystem monitoring unit is used to monitor the subsystems in the new energy grid-connected power generation system, the distribution acquisition unit is used to acquire the distribution of operating parameters, the distribution fitting unit is used to fit the distribution, and the difference comparison unit is used to acquire the information loss generated during the fitting process.

[0013] Furthermore, the comparison management module includes: a real-time sampling unit, a real-time comparison unit, and an information prompting unit. The real-time sampling unit is used to collect real-time operating data of the monitored unit and the reference unit of the monitored unit. The real-time comparison unit is used to compare the real-time data and calculate the difference judgment value. The information prompting unit is used to prompt relevant management personnel when the deviation of the difference evaluation value exceeds the threshold.

[0014] Compared with the prior art, the beneficial effects of the present invention are: This invention uses generator sets exhibiting similar changing patterns to the target generator set as reference units. This allows generator sets without direct connection to the target generator set to serve as reference units, reducing the impact of circuit connections that could disrupt the judgment of abnormal data. Such connections prevent precise identification of faulty generator sets. By analyzing the information entropy of real-time operating parameters, the invention can accurately grasp the changing trends of data from each generator set, screening potentially faulty generator sets within the same local power generation network. Further comparison with the operating parameters of the reference units, through a cross-comparison method, improves the accuracy and effectiveness of safety supervision. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the structure of an intelligent new energy grid connection status monitoring system according to the present invention; Figure 2 This is a flowchart illustrating an intelligent new energy grid connection status monitoring method according to the present invention. Figure 3 This is an example diagram of N sampling data as described in an embodiment of the present invention; Figure 4This is an example diagram of N2 sample data as described in an embodiment of the present invention; Figure 5 This is a flowchart illustrating the process of extracting the periodic variation pattern of the operating status of the first target unit according to an embodiment of the present invention; Figure 6 This is a flowchart illustrating the calculation of the difference evaluation value between the reference unit and the first target unit according to an embodiment of the present invention. Figure 7 This is a flowchart of the similarity calculation method for the changes in operating parameters of the generator set and the upstream equipment described in an embodiment of the present invention; Figure 8 This is a flowchart of the method for judging the degree of deviation of the difference evaluation value according to an embodiment of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Example: Figure 2 As shown, this invention provides an intelligent method for monitoring the grid connection status of new energy sources. The technical solution is as follows: Step S100: In a new energy grid-connected power generation system including several generator sets, a certain generator set is set as the first target unit, and a certain operating parameter of the first target unit is used as the target parameter. From the historical records of the target parameter, the periodic change pattern of the operating status of the first target unit is obtained, and the regular characteristics in the periodic change pattern are extracted. Among them, such as Figure 5 As shown, step S100 includes: Step S101: Obtain the operation record of the target parameters of the first target unit within the time period T. Set the unit sampling period to T0. After each unit sampling period, collect the target parameters once to obtain one sampled data point, denoted as... In this embodiment, an appendix can be used. Figure 3 This indicates the step.

[0018] Step S102: Obtain N consecutive sampling data points, arrange the N sampling data points according to the sampling time sequence to obtain the sampling sequence L, and form the second sampling period by N2 sampling positions, wherein the time interval between any two adjacent sampling positions is T0, and N>N2; In this embodiment, this step specifically involves processing the aforementioned sampling data. Arrange the samples chronologically to obtain the corresponding sampling sequence L, which contains the sampled data. The sampling position segment between two positions in the sampling sequence L is denoted as N2, and the sampling period corresponding to N2 is denoted as the second sampling period. (See attached...) Figure 4 As shown.

[0019] Step S103: The first sampling position of the second sampling period is associated with the nk-th sampled data in the sampling sequence L. Each sampling position in the second sampling period is associated with the sampled data in the sampling sequence in turn. The sampled data at the corresponding position in the sampling sequence L is obtained at each sampling position and is denoted as the first sampling sequence of the second sampling period. The average value of the sampled data in the first sampling sequence of the second sampling period is calculated and denoted as A21. In this embodiment, the appendix Figure 4 The data corresponding to the first sampling position in N2 corresponds to the nk-th sampling data in the sampling sequence L. Following the chronological order, the second sampling position in N2 corresponds to the (nk+1)-th sampling data. This pattern continues until all sampling positions in N2 correspond to the sampling data in the corresponding sampling sequence. The last correspondence is the N2-th sampling position corresponding to the (nk+N2)-th sampling data, thus obtaining the first sampling sequence under the second sampling period. The average value of the data under this sequence is calculated and denoted as A21.

[0020] Step S104: Record the first nk-1 sampled data in the sampling sequence L as the first sampling sequence of the first sampling period, calculate the average value of the sampled data in the first sampling sequence of the first sampling period, and record it as A11. Move the second sampling sequence by 1 unit sampling period along the time order of the sampled data in the sampling sequence, and repeat the method in step S103 to obtain the second sampling sequence of the second sampling period. Calculate the average value of the sampled data in the second sampling sequence of the second sampling period and record it as A22. Record the first nk sampled data in the sampling sequence as the second sampling sequence of the first sampling period, calculate the average value of the sampled data in the second sampling sequence of the first sampling period, and record it as A12.

[0021] In this embodiment, corresponding to the second sampling period, the first nk-1 sampled data are recorded as the first sampling sequence of the first sampling period and the average value of the corresponding data is calculated and recorded as A11. Then, the sampling period is moved forward or backward by one unit, and the average value of the corresponding sampled sequence A22 is calculated again based on the calculation method in step S103. At this time, since the sampling sequence has been moved, the corresponding first nk sampled data are also different. At this time, the average value of the corresponding data is calculated again and recorded as A12. This calculation method of changing the sampling period of the moving sampling sequence can keenly grasp the changing trend of the data of each generator set.

[0022] Step S105: Shift the first sampling sequence and the second sampling sequence by m sampling periods, match the m-th sampling sequence of the first sampling period with the m-th sampling sequence of the second sampling period, and calculate the difference between the average values ​​of the sampled data. G m, Gm=|A1m-A2m|, where A1m represents the average value of the sampled data in the m-th sampled sequence of the first sampling period, and A2m represents the average value of the sampled data in the m-th sampled sequence of the second sampling period; In other words, based on the steps above, we know that m can be 1, 2, 3, ... For example, when m=3, the corresponding average value is denoted as A13 and A23, then G3=|A13-A23|.

[0023] Step S106: Collect the differences between the averages of c sampled data points. Find the largest difference among these differences, denoted as Gmo. Obtain the position of the mo-th sample sequence in the sampling sequence for the second sampling period. The first sampled data point in the mo-th sample sequence of the second sampling period corresponds to the ko-th sampled data point in the sampling sequence L. In this embodiment, differences Gm1, Gm2, ..., Gmc are selected, and the maximum value is selected and denoted as Gmo. Using the sampling position corresponding to the maximum value as a basis, the first sampled data point in a sampling sequence is found, and the corresponding position in the sampling sequence L is denoted as the ko-th sampled data point.

[0024] Step S107: In the sampling sequence L, the total sampling time corresponding to the first ko-1 sampled data is collected to obtain the target sampling period. The time length of the target sampling period of the first target unit is denoted as T1. The unit sampling period between the ko-1 sampled data and the ko sampled data in the sampling sequence L is obtained, and the midpoint of the unit sampling period is denoted as the transition time.

[0025] In this embodiment, the total sampling time corresponding to the first ko-1 sampled data is found based on the ko-1 sampled data, and the target sampling period is obtained. The concept of transition time is defined, which is based on the unit sampling period between the ko-1 sampled data and the ko sampled data, and the midpoint of the unit sampling period is taken as the transition time.

[0026] Step S200: In the new energy grid-connected power generation system, the generator set that has the same periodic change pattern as the first target unit is recorded as the reference unit. In the operation records of the first target unit and the reference unit, the time interval in which the same pattern characteristics appear is obtained. In the same change cycle, the difference between the target parameters of the reference unit and the first generator set is used as the difference evaluation value. Preferredly, a generator set with a different upstream equipment from the first target unit is selected as a reference unit for the first target unit; Among them, such as Figure 6 As shown, step S200 includes: Step S201: In the new energy grid-connected power generation system, a generator set that is different from the first target generator set is designated as the second target generator set. The operation record of the target parameters of the second target generator set within the time period T is obtained. The steps in step S100 are repeated to obtain the target sampling period of the second target generator set. The time length of the target sampling period of the second target generator set is designated as T2. When T1=T2, the second target generator set is the reference generator set of the first target generator set. The transition time corresponding to the reference generator set is obtained. Step S202: Calculate the reference delay , Where t1 represents the transition time corresponding to the first target unit, and t2 represents the transition time corresponding to the second target unit; Step S203: Plot the function graph of the target parameters of the first target unit changing with time in the corresponding target sampling period to obtain the first parameter image; The second parameter image is obtained by plotting the function graph of the target parameters changing with time for the second target unit in the corresponding target sampling period. Furthermore, the first parameter image and the second parameter image are plotted in the same coordinate system; Step S204: In the same coordinate system, obtain the point corresponding to any sampled data in the second parameter image, obtain the shortest distance from that point to the first parameter image, and sum the shortest distances from all points corresponding to the sampled data to the first parameter image to obtain the difference evaluation value between the reference unit and the first target unit. In this embodiment, the summation is recorded as the sum of all the shortest distances.

[0027] Step S300: In the new energy grid-connected power generation system, a subsystem consisting of several generator sets and an upstream device of the generator set is obtained. Real-time operation records in the subsystem are collected. By calculating the similarity of the changes in the operating parameters of each generator set and the upstream device, the generator set with the highest correlation with the upstream device among the several generator sets is set as the monitored generator set. Among them, such as Figure 7 As shown, step S300 includes: Step S301: Obtain real-time operating data of the target parameters of a generator set in the subsystem. Within a time period of TC, obtain the real-time operating data and convert it into a data distribution column, denoted as data distribution column P1. This distribution column transforms the abstract random event pattern into an intuitive data arrangement form, which is often used to calculate expected values, analyze the probability distribution characteristics of events, and solve sampling problems in reality.

[0028] Obtain the measurement range of the target parameter in a certain generator set, and record the measurement range as the first measurement interval; In this embodiment, the measurement range is set differently based on different target parameters, i.e., different operating parameters. The operating parameters in the generator set include, but are not limited to, the following: Electrical parameters, including: Active power: Reflects the mechanical power output of the generator, directly affecting the grid load capacity. Reactive power: Used to regulate grid voltage stability; the excitation current needs to be adjusted according to the power factor. Stator voltage and stator current: Monitor the output voltage level and load current magnitude, respectively. Excitation voltage and excitation current: Control the magnetic field strength, affecting power generation efficiency and safety. Temperature parameters, including: Stator winding temperature and rotor winding temperature: Ensure equipment thermal balance and prevent overheating damage. Bearing temperature and cooling system temperature: Monitor the lubrication system and heat dissipation efficiency. Mechanical parameters, including: Speed: Maintain grid stability; needs to be adjusted in conjunction with the turbine speed control system. Rated speed: The reference speed for maintaining the frequency of a specific model, usually matched with power. System parameters, including: Hydrogen system parameters: Monitor hydrogen pressure, temperature, humidity, and hydrogen leakage. Cooling water system parameters: Check internal cooling water temperature, flow rate, water quality, etc., to ensure heat dissipation efficiency. Step S302: Obtain the upstream device connected to a certain generator set, obtain the parameter item corresponding to the target parameter in the upstream device, record the parameter item as the second target parameter, perform linear transformation or nonlinear transformation on the second target parameter, represent the real-time running data of the second target parameter in TC in the first measurement interval, convert the running data into a data distribution column, and obtain the data distribution column Q1, where the start and end times of the time period with a time length of TC are the same. The selection criteria for corresponding parameter items include that the target parameter and the second target parameter are of the same type. For example, when the target parameter is the generator voltage, the second target parameter is the generator voltage of the subsystem; when the target parameter is the generator power, the second target parameter is the generator power of the subsystem. In this embodiment, the uplink device connected to a generator set can be a power transmission device, an electrical connection device, and some auxiliary systems. The operating parameters of the corresponding uplink device are obtained. These operating parameters vary depending on the different uplink devices, but the selection criteria for the corresponding parameter items must meet the requirements of the previous section.

[0029] Step S303: Take distribution column Q1 as the reference distribution and distribution column P1 as the fitting distribution. Fit the reference distribution with the fitting distribution and calculate the information loss generated during the fitting process. The information loss is the difference in information entropy between the fitting distribution and the reference distribution. In this embodiment, the fitted probability distribution function is used to find the function that best fits the given dataset. The goal is to find a function that best fits the given dataset. Common probability distribution functions include the normal distribution, exponential distribution, and Poisson distribution. By fitting the probability distribution function, we can better understand the distribution patterns of the data, predict future data trends, and make relevant statistical inferences. In this embodiment, the distribution columns Q1 and P1 are fitted, and the difference in information entropy between the fitted distribution and the reference distribution is obtained. In information theory, the difference in information entropy is usually calculated using the relative entropy KL divergence, which measures the difference between two probability distributions. If one distribution is the true distribution and the other is the fitted distribution, the difference represents the information loss incurred when fitting the true distribution with the theoretical distribution.

[0030] A preferred approach is to obtain the information entropy difference between the two distributions by calculating the relative entropy between them, i.e., the KL divergence. This embodiment provides a specific solution, such as when a subsystem includes an uplink device U1, a generator set W1 and a generator set W2, the distribution Q1 corresponding to the second target parameter of U1 is obtained, the distribution corresponding to the target parameter of W1 is denoted as P1 and the distribution corresponding to the target parameter of W2 is denoted as P2; calculate , ,when At that time, generator set W1 was selected as the monitored unit; Step S304: Traverse the real-time operating data of all generator sets in the subsystem, obtain the information loss between the real-time operating data of each generator set and the real-time operating data of the second target parameter of the upstream equipment, and set the generator set with the minimum information loss as the monitored unit.

[0031] Step S400: Obtain the reference unit of the monitored unit, compare the real-time operating data of the monitored unit with the reference unit of the monitored unit, and when the deviation of the difference evaluation value corresponding to the real-time operating data from the difference evaluation value corresponding to the real-time operating data exceeds the threshold, issue a prompt to the relevant management personnel. Among them, such as Figure 8 As shown, step S400 includes: Step S401: Obtain the target sampling period corresponding to the controlled unit and the duration of the target sampling period; obtain the reference delay between the monitored unit and its reference unit, denoted as... The difference evaluation value between the monitored unit and the reference unit of the monitored unit. ; Step S402: From At the start of the event, real-time operation records of the target parameters of the monitored unit are collected. The collection duration is the length of the target sampling period. A graph of the real-time operation records of the target parameters as a function of time is plotted and recorded as the first operation graph. Step S403: From Starting at a certain time, real-time operational records of the target parameters of the monitored unit and the reference unit are collected. The collection duration is the length of the target sampling period. A graph of the real-time operational records of the target parameters as a function of time is plotted, denoted as the second operational graph. ; Step S404: After each unit sampling period, take a real-time sampling point on the second running image, calculate the shortest distance from each real-time sampling point to the first running image, and sum the shortest distances to obtain the current distance. ,when When this happens, a prompt is made to the generator set management personnel, where H represents the difference judgment value. Indicates obtaining and The maximum value in.

[0032] Therefore, by processing relevant data from generator sets, generator sets with the same changing patterns as reference generator sets, upstream equipment connected to generator sets, and monitored generator sets, this embodiment can accurately analyze the changing trends of data from each generator set, thereby accurately predicting and monitoring potential faults in the power generation system in real time.

[0033] To better implement the above methods, an intelligent new energy grid connection status monitoring system is also proposed, such as... Figure 1 As shown, the system includes: a change pattern management module, a difference evaluation value management module, a real-time monitoring module, and a comparison management module; The variation pattern management module is used to collect historical data to manage the periodic variation pattern of the generator set. The variation pattern management module includes: a data acquisition unit, a sampling period management unit, a sampling movement unit, a difference comparison unit, and a transition time extraction unit. The data acquisition unit is used to collect the operating data of the generator set, the sampling period management unit is used to obtain the second sampling period, the sampling movement unit is used to move the second sampling period in the sampling sequence, the difference comparison unit is used to calculate the difference between the average values ​​of the sampled data, and the transition time extraction unit is used to obtain the transition time. The difference evaluation value management module is used to compare the operating parameters of the reference unit and the first target unit and calculate the difference of the target parameters. The difference evaluation value management module includes a unit acquisition unit, a time delay calculation unit and a difference evaluation unit. The unit acquisition unit is used to acquire the reference unit of the first target unit, the time delay calculation unit is used to calculate the time difference in the transition time, and the difference evaluation unit is used to acquire the first parameter image and the second parameter image and calculate the difference evaluation value between the reference unit and the first target unit. The real-time monitoring module is used to manage the real-time operating status of the generator set. The real-time monitoring module includes: a subsystem monitoring unit, a distribution column acquisition unit, a distribution column fitting unit, and a difference comparison unit. The subsystem monitoring unit is used to monitor the subsystems in the new energy grid-connected power generation system. The distribution column acquisition unit is used to acquire the distribution column of operating parameters. The distribution column fitting unit is used to fit the distribution column. The difference comparison unit is used to acquire the information loss generated during the fitting process. The comparison management module is used to compare the real-time operating data of the monitored unit with that of the reference unit. The real-time monitoring module includes a subsystem monitoring unit, a distribution column acquisition unit, a distribution column fitting unit, and a difference comparison unit. The subsystem monitoring unit is used to monitor the subsystems in the new energy grid-connected power generation system. The distribution column acquisition unit is used to acquire the distribution column of operating parameters. The distribution column fitting unit is used to fit the distribution column. The difference comparison unit is used to acquire the information loss generated during the fitting process.

[0034] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. An intelligent method for monitoring the grid connection status of new energy sources, characterized in that: The method includes the following steps: Step S100: In a new energy grid-connected power generation system including several generator sets, a certain generator set is set as the first target unit, and a certain operating parameter of the first target unit is used as the target parameter. From the historical records of the target parameter, the periodic change pattern of the operating status of the first target unit is obtained, and the regular characteristics in the periodic change pattern are extracted. Step S200: In the new energy grid-connected power generation system, the generator set that has the same periodic change pattern as the first target unit is recorded as the reference unit. In the operation records of the first target unit and the reference unit, the time interval in which the same pattern characteristics appear is obtained. In the same change cycle, the difference between the target parameters of the reference unit and the first generator set is used as the difference evaluation value. Step S200 includes: Step S201: In the new energy grid-connected power generation system, a generator set that is different from the first target generator set is designated as the second target generator set. The operation record of the target parameters of the second target generator set within the time period T is obtained. The steps in step S100 are repeated to obtain the target sampling period of the second target generator set. The time length of the target sampling period of the second target generator set is denoted as T2. When T1=T2, the second target generator set is the reference generator set of the first target generator set. The transition time corresponding to the reference generator set is obtained. Step S202: Calculate the reference delay , Where t1 represents the transition time corresponding to the first target unit, and t2 represents the transition time corresponding to the second target unit; Step S203: Plot the function graph of the target parameters changing with time for the first target unit in the corresponding target sampling period to obtain the first parameter image; plot the function graph of the target parameters changing with time for the second target unit in the corresponding target sampling period to obtain the second parameter image; plot the first parameter image and the second parameter image in the same coordinate system. Step S204: In the same coordinate system, obtain the point corresponding to any sampled data in the second parameter image, obtain the shortest distance from the point to the first parameter image, sum the shortest distances from all sampled data points to the first parameter image, and obtain the difference evaluation value between the reference unit and the first target unit. Step S300: In the new energy grid-connected power generation system, a subsystem consisting of several generator sets and an upstream device of the generator sets is obtained. Real-time operation records in the subsystem are collected. By calculating the similarity of the changes in the operating parameters of each generator set and the upstream device, the generator set with the highest correlation with the upstream device among the several generator sets is set as the monitored generator set. Step S300 includes: Step S301: Obtain real-time operating data of the target parameters of a certain generator set in the subsystem. During a time period of TC, obtain the real-time operating data, convert the real-time operating data into a data distribution column, denoted as data distribution column P1, obtain the measurement range of the target parameters in the certain generator set, and denot the measurement range as the first measurement interval. Step S302: Obtain the uplink device connected to the generator set, obtain the parameter item corresponding to the target parameter in the uplink device, record the parameter item as the second target parameter, perform linear transformation or nonlinear transformation on the second target parameter, represent the real-time running data of the second target parameter in the time length TC in the first measurement interval, convert the running data into a data distribution column to obtain the data distribution column Q1, wherein the start and end times of the time period with the time length TC are the same; Step S303: Take distribution column Q1 as the reference distribution and distribution column P1 as the fitting distribution. Fit the reference distribution with the fitting distribution and calculate the information loss generated during the fitting process. The information loss is the difference in information entropy between the fitting distribution and the reference distribution. Step S304: Traverse the real-time operating data of all generator sets in the subsystem, obtain the information loss between the real-time operating data of each generator set and the real-time operating data of the second target parameter of the uplink device, and set the generator set with the minimum information loss as the monitored generator set. Step S400: Obtain the reference unit of the monitored unit, compare the real-time operating data of the monitored unit with the reference unit of the monitored unit, and when the deviation of the difference evaluation value corresponding to the real-time operating data of the monitored unit and the reference unit of the monitored unit exceeds the threshold, issue a prompt to the relevant management personnel.

2. The intelligent new energy grid connection status monitoring method according to claim 1, characterized in that: Step S100 includes: Step S101: Obtain the operation record of the target parameters of the first target unit within the time period T, set the unit sampling period to T0, and collect the target parameters once after each unit sampling period to obtain one sample data. Step S102: Obtain N consecutive sampling data points, arrange the N sampling data points according to the sampling time sequence to obtain a sampling sequence L, and form a second sampling period by N2 sampling positions, wherein the time interval between any two adjacent sampling positions is T0, and N>N2; Step S103: The first sampling position of the second sampling period is associated with the nk-th sampled data in the sampling sequence L. Each sampling position in the second sampling period is associated with the sampled data in the sampling sequence in turn. The sampled data at the corresponding position in the sampling sequence L is obtained at each sampling position and is denoted as the first sampling sequence of the second sampling period. The average value of the sampled data in the first sampling sequence of the second sampling period is calculated and denoted as A21. Step S104: Record the first nk-1 sampled data in the sampling sequence L as the first sampling sequence of the first sampling period, calculate the average value of the sampled data in the first sampling sequence of the first sampling period, and record it as A11. Move the second sampling sequence by 1 unit sampling period along the time order of the sampled data in the sampling sequence, and repeat the method in step S103 to obtain the second sampling sequence of the second sampling period. Calculate the average value of the sampled data in the second sampling sequence of the second sampling period and record it as A22. Record the first nk sampled data in the sampling sequence as the second sampling sequence of the first sampling period, calculate the average value of the sampled data in the second sampling sequence of the first sampling period, and record it as A12.

3. The intelligent new energy grid connection status monitoring method according to claim 2, characterized in that: Step S100, the step of extracting the regularity features from the periodic variation pattern, includes: Step S105: Match the m-th sampling sequence of the first sampling period with the m-th sampling sequence of the second sampling period, and calculate the difference Gm between the average values ​​of the sampled data, Gm=|A1m-A2m|, where A1m represents the average value of the sampled data in the m-th sampling sequence of the first sampling period, and A2m represents the average value of the sampled data in the m-th sampling sequence of the second sampling period; Step S106: Collect the difference between the average values ​​of c sampled data, obtain the largest difference from the differences between the average values ​​of the c sampled data, denoted as Gmo, and obtain the position of the mo-th sampled sequence in the sampling sequence of the second sampling period, wherein the first sampled data in the mo-th sampled sequence of the second sampling period corresponds to the ko-th sampled data in the sampling sequence L. Step S107: In the sampling sequence L, the total sampling time corresponding to the first ko-1 sampled data is collected to obtain the target sampling period. The time length of the target sampling period of the first target unit is denoted as T1. The unit sampling period between the ko-1th sampled data and the koth sampled data in the sampling sequence L is obtained, and the midpoint of the time of the unit sampling period is denoted as the transition time.

4. The intelligent new energy grid connection status monitoring method according to claim 3, characterized in that: Step S400 includes: Step S401: Obtain the target sampling period corresponding to the controlled unit and the time length of the target sampling period; obtain the reference delay between the monitored unit and the reference unit of the monitored unit, denoted as... The difference evaluation value d0 between the monitored unit and the reference unit of the monitored unit; Step S402: From At a certain moment, real-time operation records of the target parameters of the monitored unit are collected. The collection duration is the time length of the target sampling period. A function graph of the real-time operation records of the target parameters with respect to time is plotted and recorded as the first operation graph. Step S403: From Starting at a certain time, real-time operation records of the target parameters of the monitored unit and the reference unit are collected. The collection duration is the length of the target sampling period. A graph of the real-time operation records of the target parameters as a function of time is plotted and denoted as the second operation graph. ; Step S404: After each unit sampling period, take a real-time sampling point on the second running image, calculate the shortest distance from each real-time sampling point to the first running image, and sum the shortest distances to obtain the current distance. ,when At that time, a reminder should be given to the generator set management personnel, among which, Indicates the difference judgment value. Used to indicate acquisition and The maximum value in.

5. An intelligent new energy grid connection status monitoring system, used to execute the intelligent new energy grid connection status monitoring method according to any one of claims 1-4, characterized in that: The system includes: The system comprises a change pattern management module, a difference evaluation value management module, a real-time monitoring module, and a comparison management module. The change pattern management module is used to collect historical data to manage the periodic change patterns of the generator set. The difference evaluation value management module is used to compare the operating parameters of the reference unit and the first target unit and calculate the difference in the target parameters. The real-time monitoring module is used to manage the real-time operating status of the generator set. The comparison management module is used to compare the real-time operating data of the monitored unit with the reference unit of the monitored unit.

6. The intelligent new energy grid connection status monitoring system according to claim 5, characterized in that: The change pattern management module includes: a data acquisition unit, a sampling period management unit, a sampling movement unit, a difference comparison unit, and a transition time extraction unit. The data acquisition unit is used to collect the operating data of the generator set, the sampling period management unit is used to obtain the second sampling period, the sampling movement unit is used to move the second sampling period in the sampling sequence, the difference comparison unit is used to calculate the difference between the average values ​​of the sampled data, and the transition time extraction unit is used to obtain the transition time.

7. The intelligent new energy grid connection status monitoring system according to claim 6, characterized in that: The difference evaluation value management module includes: a unit acquisition unit, a time delay calculation unit, and a difference evaluation unit. The unit acquisition unit is used to acquire the reference unit of the first target unit, the time delay calculation unit is used to calculate the time difference in the transition time, and the difference evaluation unit is used to acquire the first parameter image and the second parameter image, and calculate the difference evaluation value between the reference unit and the first target unit. The real-time monitoring module includes: a subsystem monitoring unit, a distribution column acquisition unit, a distribution column fitting unit, and a difference comparison unit. The subsystem monitoring unit is used to monitor the subsystems in the new energy grid-connected power generation system, the distribution column acquisition unit is used to acquire the distribution column of operating parameters, the distribution column fitting unit is used to fit the distribution column, and the difference comparison unit is used to acquire the information loss generated during the fitting process.

8. The intelligent new energy grid connection status monitoring system according to claim 7, characterized in that: The comparison management module includes a real-time sampling unit, a real-time comparison unit, and an information prompting unit. The real-time sampling unit is used to collect real-time operating data of the monitored unit and its reference unit. The real-time comparison unit is used to compare the real-time data and calculate the difference judgment value. The information prompting unit is used to prompt relevant management personnel when the deviation of the difference evaluation value exceeds the threshold.

Images