An electric-thermal conversion rate control method based on balance capability prediction

Abstract

This application discloses an electrothermal conversion rate control method based on balance capability prediction, comprising: acquiring historical operation logs; analyzing historical operation data and historical balance capability indices in each historical operation log to determine characteristic operation data; constructing a balance capability prediction model for a multi-energy system based on the characteristic operation data and historical balance capability indices and generating a balance capability prediction index; analyzing the balance capability prediction index and real-time electrothermal conversion rate according to a preset control analysis model to determine an initial control strategy; setting a verification period; obtaining the actual application degree of the initial control strategy within the verification period; and determining whether to generate a correction instruction for the initial control strategy based on the actual balance capability index and the actual application degree, thereby improving the accuracy of balance capability prediction and the control effect of the electrothermal conversion rate control method, and ensuring the stable operation of the multi-energy system.

Description

Technical Field

[0001] This application relates to the field of balance capacity prediction technology, and in particular to an electrothermal conversion rate control method based on balance capacity prediction. Background Technology

[0002] Multi-energy energy balance capability prediction refers to the prediction of the balance state of multiple energy sources over a period of time by analyzing the supply and demand of multiple energy sources, thereby providing a scientific basis for energy planning and decision-making, reducing energy waste and ensuring the stable operation of multi-energy systems.

[0003] In the existing technology, the accuracy of balance capability prediction is low, resulting in poor control effect of subsequent electrothermal conversion rate control methods. Therefore, how to improve the accuracy of balance capability prediction and the control effect of electrothermal conversion rate control methods is a technical problem that needs to be solved. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides an electrothermal conversion rate control method based on balance capability prediction. By analyzing historical operating data and historical balance capability indices, characteristic operating data is determined, and a balance capability prediction model is constructed to obtain the balance capability prediction index for future periods. Based on a preset control analysis model, the balance capability prediction index and real-time electrothermal rotation speed are analyzed to determine an initial control strategy. Whether to generate a correction command is determined based on the actual balance capability index and the practical application of the initial control strategy. This improves the accuracy of balance capability prediction and the control effect of the electrothermal conversion rate control method, ensuring the stable operation of the multi-energy system.

[0005] In some embodiments of this application, an electrothermal conversion rate control method based on balance capability prediction is provided, including:

[0006] Obtain historical operation logs of multi-energy systems, analyze the historical operation data and historical balance capacity index in each historical operation log, and determine characteristic operation data;

[0007] A balance capacity prediction model for a multi-energy system is constructed based on characteristic operating data and historical balance capacity indices. A balance capacity prediction index for future periods is then generated based on the balance capacity prediction model.

[0008] Based on the preset control analysis model, the balance capacity prediction index and real-time electrothermal conversion rate are analyzed to determine the initial control strategy, and control commands for the real-time electrothermal conversion rate are generated based on the initial control strategy.

[0009] Set the test duration, obtain the actual application degree of the initial control strategy within the test duration, and determine whether to generate a correction instruction for the initial control strategy based on the actual balance capability index and the actual application degree.

[0010] In some embodiments of this application, historical operation data and historical balancing capacity index in each historical operation log are analyzed to determine characteristic operation data, including:

[0011] Acquire historical operation logs of multiple historical monitoring cycles of the multi-energy system, and determine the historical balance capacity index at each acquisition time node based on a preset balance capacity index algorithm for each historical operation log.

[0012] Using the historical monitoring duration of each historical monitoring cycle as the time reference line and the preset time interval as the collection time node, the historical operation data and historical balance capacity index in the historical operation log of the corresponding historical monitoring cycle are obtained according to the collection time node and mapped to the corresponding time reference line to obtain the impact analysis diagram of each historical monitoring cycle.

[0013] For each impact analysis diagram, the historical balance capability index of all collection time nodes is analyzed for fluctuation. Collection time nodes with fluctuation of historical balance capability index greater than the preset fluctuation threshold are selected. The selected adjacent collection time nodes are set as a characteristic fluctuation period, and the period to be analyzed for each characteristic fluctuation period is determined.

[0014] Obtain historical operational data for each time period to be analyzed in the same impact analysis diagram, and determine whether the degree of change of each historical operational data for the time period to be analyzed is greater than the preset degree of change threshold. If not, remove the corresponding historical operational data; if so, retain the corresponding historical operational data.

[0015] Obtain the first fluctuation feature of the retained historical operating data for each period to be analyzed and the second fluctuation feature of the historical balance capability index for the corresponding period of characteristic fluctuation. Determine the characteristic operating data based on the first and second fluctuation features.

[0016] In some embodiments of this application, determining characteristic operation data based on a first fluctuation characteristic and a second fluctuation characteristic includes:

[0017] Determine the initial mutation time node of the retained historical running data in the characteristic period to be analyzed, and obtain the first fluctuation feature of each retained historical running data in the initial mutation time node and the characteristic period to be analyzed. The first fluctuation feature includes the fluctuation value, fluctuation speed and fluctuation trend of the corresponding historical running data at multiple preset analysis time nodes in the initial mutation time node and the characteristic period to be analyzed.

[0018] Obtain the second fluctuation feature of the historical balance capability index for each characteristic fluctuation period corresponding to the characteristic fluctuation period to be analyzed. The second fluctuation feature includes the fluctuation value, fluctuation rate and fluctuation trend of the historical balance capability index at multiple preset analysis time nodes in the corresponding characteristic fluctuation period.

[0019] The second fluctuation feature of the historical balance capability index during the characteristic fluctuation period is compared with the first fluctuation feature of each historical operating data of the period to be analyzed corresponding to the characteristic fluctuation period. Historical operating data with a similarity greater than the preset similarity threshold are selected and set as the operating data of interest for the corresponding characteristic fluctuation period.

[0020] Pre-set a preset balance capability index range to determine the preset balance capability index range in which the historical balance capability index of all characteristic fluctuation periods in different influence analysis charts falls;

[0021] By comparing the operational data of each characteristic fluctuation period within the same preset balance capability index range, the frequency of occurrence of the same operational data of each characteristic fluctuation period within the same preset balance capability index range can be obtained.

[0022] If the frequency of occurrence is less than the preset frequency threshold, the corresponding data of interest is removed. If the frequency of occurrence is greater than the preset frequency threshold, the corresponding data of interest is retained and the retained data of interest is set as feature data.

[0023] In some embodiments of this application, the fluctuation trend of each historical operating data in the period to be analyzed at all preset analysis time nodes is compared with the fluctuation trend of the historical balance capability index of the characteristic fluctuation period at all preset analysis time nodes to obtain the degree of deviation between the fluctuation trend of each historical operating data and the historical balance capability index.

[0024] Based on the principle of time node comparison, each historical operating data in the period to be analyzed is matched with the preset analysis time node of the historical balance ability index of the characteristic fluctuation period to obtain multiple preset analysis time node groups.

[0025] Obtain the fluctuation values ​​of the first fluctuation feature and the second fluctuation feature of each preset analysis time node group and calculate the difference to obtain the fluctuation value difference of each preset analysis time node group;

[0026] The fluctuation velocities of the first and second fluctuation characteristics of each preset analysis time node group are obtained and subtracted to obtain the fluctuation velocity difference value of each preset analysis time node group;

[0027] The similarity between each historical operating data point and the corresponding historical balance capability index is generated based on the degree of deviation in the fluctuation trend, the difference in the fluctuation value of each preset analysis time node group, and the difference in the fluctuation speed.

[0028] The formula for calculating the similarity is:

[0029]

[0030] Where D is the similarity, x1 is the first similarity conversion coefficient, q1 is the weight coefficient of the fluctuation trend, b1 is the degree of deviation of the fluctuation trend, x2 is the second similarity conversion coefficient, q2 is the weight coefficient of the fluctuation value, b2i is the difference of the fluctuation value of the i-th preset analysis time node group, m is the total number of preset analysis time node groups, x3 is the third similarity conversion coefficient, q3 is the weight coefficient of the fluctuation speed, and b3i is the difference of the fluctuation speed of the i-th preset analysis time node group.

[0031] In some embodiments of this application, a balance capacity prediction model for a multi-energy system is constructed based on characteristic operating data and historical balance capacity indices, including:

[0032] Obtain the second fluctuation characteristics of the historical balance capacity index for all characteristic fluctuation periods, and the first fluctuation characteristics of the characteristic operation data of the period to be analyzed corresponding to the characteristic fluctuation periods;

[0033] Based on the same characteristic running data and the first fluctuation feature of the same characteristic running data for all the periods to be analyzed, the data is divided into training input data and test input data. The historical balance ability index of the characteristic fluctuation period corresponding to the period to be analyzed in the training input data and the second fluctuation feature of the historical balance ability index are used as training output data to obtain the initial balance ability prediction model of each characteristic running data and the historical balance ability index.

[0034] The credibility of the corresponding initial balance capability prediction model is generated based on the test input data. The initial balance capability prediction model with a credibility of less than the preset credibility threshold is iteratively trained until the credibility is greater than the preset credibility threshold.

[0035] The fusion coefficient of each initial equilibrium capacity prediction model is set according to the credibility of each initial equilibrium capacity prediction model;

[0036] All initial balance capacity prediction models and their corresponding fusion coefficients are fused, and the fused model is set as the balance capacity prediction model for a multi-energy system.

[0037] In some embodiments of this application, generating a future time period balance capacity prediction index based on a balance capacity prediction model includes:

[0038] Acquire real-time characteristic operation data of multi-energy system during preset monitoring period, analyze the real-time characteristic operation data during preset monitoring period, and determine the real-time fluctuation characteristics of real-time characteristic operation data during preset monitoring period.

[0039] All real-time feature operation data and the corresponding real-time fluctuation features of the real-time feature operation data are input into the balance capacity prediction model to obtain the balance capacity prediction index for future periods.

[0040] In some embodiments of this application, an initial control strategy is determined by analyzing the balance capability prediction index and the real-time electrothermal conversion rate based on a preset control analysis model, including:

[0041] The current balance capability prediction index is analyzed based on the preset control analysis model to obtain the preset control strategy library corresponding to the current balance capability prediction index.

[0042] The preset control strategy library includes several preset electrothermal conversion rate differences of the current balance capability prediction index, and each preset electrothermal conversion rate difference is associated with a specific preset control strategy.

[0043] The real-time electrothermal conversion rate is obtained, and the difference between the real-time electrothermal conversion rate and the standard electrothermal conversion rate corresponding to the current balance capacity prediction index is calculated to obtain the real-time electrothermal conversion rate difference.

[0044] A similarity analysis is performed between the real-time electrothermal conversion rate difference and several preset electrothermal conversion rate differences in the preset control strategy library. The preset control strategy corresponding to the preset electrothermal conversion rate difference with the highest similarity is set as the initial control strategy for the real-time electrothermal conversion rate.

[0045] In some embodiments of this application, obtaining the actual application degree of the initial control strategy within the testing period includes:

[0046] Multiple application evaluation metrics are pre-defined;

[0047] Obtain real-time operational data within the testing period, and determine the associated operational data for each application evaluation indicator based on the correlation between the application evaluation indicators and the real-time operational data;

[0048] The associated operational data is compared with the standard operational data of the corresponding application evaluation indicators to obtain the first degree of deviation between the associated operational data and the standard operational data. Based on the first degree of deviation, the sub-application degree of the associated operational data is generated.

[0049] The second degree of deviation between the associated operating data and the standard operating data before the initial control strategy is obtained, and the trend and rate of change of the deviation degree from the first degree of deviation of the associated operating data are generated.

[0050] The first compensation coefficient for the sub-application degree of the corresponding associated operational data is generated based on the trend of change; the second compensation coefficient for the sub-application degree of the corresponding associated operational data is generated based on the rate of change.

[0051] The actual application degree of the initial control strategy is generated based on the sub-application degree of the associated operational data, the first compensation coefficient, the second compensation coefficient, and the weight coefficients of the corresponding application evaluation indicators.

[0052] The formula for calculating the actual application degree is:

[0053]

[0054] Where Y represents the actual application degree, f1 is the weight coefficient of the first application evaluation indicator, n1 is the total number of associated operational data for the first application evaluation indicator, and Z1 s1 c11 is the sub-application degree of the s1th associated running data of the first application evaluation index. s1 c21 is the first compensation coefficient for the s1th associated operational data of the first application evaluation index. s1 Zr is the second compensation coefficient for the s1th associated operational data of the first application evaluation index. s1 Let c1r be the sub-application degree of the sr-th associated runtime data for the r-th application evaluation index. sr c2r is the first compensation coefficient for the sr-th associated operational data of the r-th application evaluation index. sr is the second compensation coefficient for the sr-th associated running data of the r-th application evaluation index, and nr is the total number of associated running data of the r-th application evaluation index.

[0055] In some embodiments of this application, determining whether to generate a correction instruction for the initial control strategy based on the actual balance capability index and actual application degree includes:

[0056] Pre-set the exponential difference threshold;

[0057] Obtain the index difference between the actual balance capability index and the predicted balance capability index at the same time point. If the index difference is less than the index difference threshold, do not send a correction instruction. If the index difference is greater than the index difference threshold, send the first-level correction instruction of the initial control strategy.

[0058] Pre-set application threshold;

[0059] If the actual application degree is greater than the application degree threshold, no correction instruction will be sent;

[0060] If the actual application degree is less than the application degree threshold, send a secondary correction instruction for the initial control strategy.

[0061] The electrothermal conversion rate control method based on balance capability prediction in this application has the following advantages compared with the prior art:

[0062] By analyzing historical operating data and historical balance capacity indices, characteristic operating data are identified, and a balance capacity prediction model is constructed to obtain the balance capacity prediction index for future periods. Based on a preset control analysis model, the balance capacity prediction index and real-time electrothermal speed are analyzed to determine the initial control strategy. Based on the actual balance capacity index and the actual application of the initial control strategy, it is determined whether to generate a correction command, thereby improving the accuracy of balance capacity prediction and the control effect of the electrothermal conversion rate control method, and ensuring the stable operation of the multi-energy system. Attached Figure Description

[0063] Figure 1 This is a flowchart illustrating a preferred embodiment of an electrothermal conversion rate control method based on balance capability prediction. Detailed Implementation

[0064] The specific embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate this application, but are not intended to limit the scope of this application.

[0065] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0066] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "multiple" means two or more.

[0067] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linkage" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0068] like Figure 1 As shown, a preferred embodiment of the present application provides a method for controlling the electrothermal conversion rate based on balance capability prediction, comprising:

[0069] Step S101: Obtain the historical operation logs of the multi-energy system, analyze the historical operation data and historical balance capacity index in each historical operation log, and determine the characteristic operation data;

[0070] Step S102: Construct a balance capacity prediction model for a multi-energy system based on characteristic operating data and historical balance capacity index, and generate a balance capacity prediction index for future periods based on the balance capacity prediction model;

[0071] Step S103: Analyze the balance capacity prediction index and real-time electrothermal conversion rate according to the preset control analysis model, determine the initial control strategy, and generate control commands for the real-time electrothermal conversion rate based on the initial control strategy.

[0072] Step S104: Set the test duration, obtain the actual application degree of the initial control strategy within the test duration, and determine whether to generate a correction instruction for the initial control strategy based on the actual balance capability index and the actual application degree.

[0073] In some embodiments of this application, historical operation data and historical balancing capacity index in each historical operation log are analyzed to determine characteristic operation data, including:

[0074] Acquire historical operation logs of multiple historical monitoring cycles of the multi-energy system, and determine the historical balance capacity index at each acquisition time node based on a preset balance capacity index algorithm for each historical operation log.

[0075] Using the historical monitoring duration of each historical monitoring cycle as the time reference line and the preset time interval as the collection time node, the historical operation data and historical balance capacity index in the historical operation log of the corresponding historical monitoring cycle are obtained according to the collection time node and mapped to the corresponding time reference line to obtain the impact analysis diagram of each historical monitoring cycle.

[0076] For each impact analysis diagram, the historical balance capability index of all collection time nodes is analyzed for fluctuation. Collection time nodes with fluctuation of historical balance capability index greater than the preset fluctuation threshold are selected. The selected adjacent collection time nodes are set as a characteristic fluctuation period, and the period to be analyzed for each characteristic fluctuation period is determined.

[0077] Obtain historical operational data for each time period to be analyzed in the same impact analysis diagram, and determine whether the degree of change of each historical operational data for the time period to be analyzed is greater than the preset degree of change threshold. If not, remove the corresponding historical operational data; if so, retain the corresponding historical operational data.

[0078] Obtain the first fluctuation feature of the retained historical operating data for each period to be analyzed and the second fluctuation feature of the historical balance capability index for the corresponding period of characteristic fluctuation. Determine the characteristic operating data based on the first and second fluctuation features.

[0079] In this embodiment, the preset balance capability index algorithm is determined based on multiple methods for evaluating the balance capability of multiple energy sources.

[0080] In this embodiment, the time period to be analyzed refers to the preset time period adjacent to each characteristic fluctuation period, wherein the preset time period is 30 minutes.

[0081] In this embodiment, it is determined whether the degree of change in the historical operating data within the period to be analyzed is greater than a preset degree of change threshold. This allows for the elimination of operating data that has no impact on the historical balance capability index, reducing the amount of data processing and analysis, and improving the speed of subsequent characteristic operating data determination and the accuracy of the balance capability prediction model.

[0082] In some embodiments of this application, determining characteristic operation data based on a first fluctuation characteristic and a second fluctuation characteristic includes:

[0083] Determine the initial mutation time node of the retained historical running data in the characteristic period to be analyzed, and obtain the first fluctuation feature of each retained historical running data in the initial mutation time node and the characteristic period to be analyzed. The first fluctuation feature includes the fluctuation value, fluctuation speed and fluctuation trend of the corresponding historical running data at multiple preset analysis time nodes in the initial mutation time node and the characteristic period to be analyzed.

[0084] Obtain the second fluctuation feature of the historical balance capability index for each characteristic fluctuation period corresponding to the characteristic fluctuation period to be analyzed. The second fluctuation feature includes the fluctuation value, fluctuation rate and fluctuation trend of the historical balance capability index at multiple preset analysis time nodes in the corresponding characteristic fluctuation period.

[0085] The second fluctuation feature of the historical balance capability index during the characteristic fluctuation period is compared with the first fluctuation feature of each historical operating data of the period to be analyzed corresponding to the characteristic fluctuation period. Historical operating data with a similarity greater than the preset similarity threshold are selected and set as the operating data of interest for the corresponding characteristic fluctuation period.

[0086] Pre-set a preset balance capability index range to determine the preset balance capability index range in which the historical balance capability index of all characteristic fluctuation periods in different influence analysis charts falls;

[0087] By comparing the operational data of each characteristic fluctuation period within the same preset balance capability index range, the frequency of occurrence of the same operational data of each characteristic fluctuation period within the same preset balance capability index range can be obtained.

[0088] If the frequency of occurrence is less than the preset frequency threshold, the corresponding data of interest is removed. If the frequency of occurrence is greater than the preset frequency threshold, the corresponding data of interest is retained and the retained data of interest is set as feature data.

[0089] In this embodiment, the frequency of occurrence = the number of times the same monitored operational data occurs in each characteristic fluctuation period within the same preset balance capability index range / the total number of characteristic fluctuation periods within the same preset balance capability index range.

[0090] In this embodiment, the preset analysis time node refers to the time node with the same time interval set in advance in the period to be analyzed and the characteristic fluctuation period. The preset analysis time node in each period to be analyzed is set according to the initial mutation node of each historical running data, while the preset analysis time node in the characteristic fluctuation period is set from the beginning.

[0091] In this embodiment, the preset balance capability index interval in which the historical balance capability index of the characteristic fluctuation period is located means that both the maximum and minimum historical balance capability index of the characteristic fluctuation period are within the corresponding preset balance capability index interval. By dividing the historical balance capability index of the characteristic fluctuation period into the corresponding preset balance capability index interval, the monitoring operation data of multiple characteristic fluctuation periods are comprehensively analyzed to improve the accuracy of subsequent characteristic operation data, thereby laying the foundation for the subsequent construction of a balance capability prediction model.

[0092] In this embodiment, by comparing the fluctuation characteristics of the historical balance capability index and each historical operating data at multiple preset analysis time nodes, the relationship between the historical balance capability index and the historical operating data at the corresponding preset analysis time nodes is obtained, thereby determining the operating data of interest. The operating data of interest in all characteristic fluctuation periods belonging to the same preset balance capability index interval are compared to obtain the frequency of occurrence of the operating data of interest in the corresponding preset balance capability index interval. The operating data of interest with low occurrence frequency is eliminated, and the remaining operating data of interest is retained and set as characteristic operating data, thereby improving the accuracy of characteristic operating data.

[0093] In this embodiment, characteristic operating data refers to data that causes significant fluctuations in the balance capability index. Improving the accuracy of characteristic operating data and constructing a balance capability prediction model based on the fluctuation characteristics of characteristic operating data and historical balance capability index improves the accuracy of the balance capability prediction model, thereby enhancing the control efficiency of the subsequent control method for determining the electrothermal conversion rate and ensuring the stable operation of the multi-energy system.

[0094] In some embodiments of this application, the fluctuation trend of each historical operating data in the period to be analyzed at all preset analysis time nodes is compared with the fluctuation trend of the historical balance capability index of the characteristic fluctuation period at all preset analysis time nodes to obtain the degree of deviation between the fluctuation trend of each historical operating data and the historical balance capability index.

[0095] Based on the principle of time node comparison, each historical operating data in the period to be analyzed is matched with the preset analysis time node of the historical balance ability index of the characteristic fluctuation period to obtain multiple preset analysis time node groups.

[0096] Obtain the fluctuation values ​​of the first fluctuation feature and the second fluctuation feature of each preset analysis time node group and calculate the difference to obtain the fluctuation value difference of each preset analysis time node group;

[0097] The fluctuation velocities of the first and second fluctuation characteristics of each preset analysis time node group are obtained and subtracted to obtain the fluctuation velocity difference value of each preset analysis time node group;

[0098] The similarity between each historical operating data point and the corresponding historical balance capability index is generated based on the degree of deviation in the fluctuation trend, the difference in the fluctuation value of each preset analysis time node group, and the difference in the fluctuation speed.

[0099] The formula for calculating the similarity is:

[0100]

[0101] Where D is the similarity, x1 is the first similarity conversion coefficient, q1 is the weight coefficient of the fluctuation trend, b1 is the degree of deviation of the fluctuation trend, x2 is the second similarity conversion coefficient, q2 is the weight coefficient of the fluctuation value, b2i is the difference of the fluctuation value of the i-th preset analysis time node group, m is the total number of preset analysis time node groups, x3 is the third similarity conversion coefficient, q3 is the weight coefficient of the fluctuation speed, and b3i is the difference of the fluctuation speed of the i-th preset analysis time node group.

[0102] In this embodiment, the time node comparison principle refers to comparing time nodes with the same time interval between the initial time node and the last time node, thereby obtaining multiple matching preset analysis time node groups. Each preset analysis time node group consists of a preset analysis time node in the period to be analyzed and a preset analysis time node corresponding to the characteristic fluctuation period.

[0103] In this embodiment, the degree of fluctuation trend deviation refers to the degree of deviation between the fluctuation trend of the first fluctuation feature and the second fluctuation feature and the magnitude of the fluctuation. The smaller the degree of fluctuation trend deviation, the smaller the difference in fluctuation magnitude and the smaller the difference in fluctuation speed, the greater the similarity.

[0104] In this embodiment, by comparing the fluctuation characteristics of historical operating data and the corresponding historical balance capability index, the similarity of the fluctuations of historical operating data and historical balance capability index in the period to be analyzed and the characteristic fluctuation period is determined, thereby accurately screening out the characteristic operating data and laying the foundation for the subsequent construction of the balance capability prediction model.

[0105] In some embodiments of this application, a balance capacity prediction model for a multi-energy system is constructed based on characteristic operating data and historical balance capacity indices, including:

[0106] Obtain the second fluctuation characteristics of the historical balance capacity index for all characteristic fluctuation periods, and the first fluctuation characteristics of the characteristic operation data of the period to be analyzed corresponding to the characteristic fluctuation periods;

[0107] Based on the same characteristic running data and the first fluctuation feature of the same characteristic running data for all the periods to be analyzed, the data is divided into training input data and test input data. The historical balance ability index of the characteristic fluctuation period corresponding to the period to be analyzed in the training input data and the second fluctuation feature of the historical balance ability index are used as training output data to obtain the initial balance ability prediction model of each characteristic running data and the historical balance ability index.

[0108] The credibility of the corresponding initial balance capability prediction model is generated based on the test input data. The initial balance capability prediction model with a credibility of less than the preset credibility threshold is iteratively trained until the credibility is greater than the preset credibility threshold.

[0109] The fusion coefficient of each initial equilibrium capacity prediction model is set according to the credibility of each initial equilibrium capacity prediction model;

[0110] All initial balance capacity prediction models and their corresponding fusion coefficients are fused, and the fused model is set as the balance capacity prediction model for a multi-energy system.

[0111] In this embodiment, the lag time refers to the length of time it takes for a sudden change in the characteristic running data to cause a sudden change in the historical balance capability index. The shorter the lag time, the faster the corresponding characteristic running data affects the historical balance capability index, and vice versa.

[0112] In this embodiment, model fusion of all initial equilibrium capacity prediction models and their corresponding weight coefficients refers to combining the capabilities of different initial equilibrium capacity prediction models to compensate for the biases and errors of individual initial equilibrium capacity prediction models. Reliability refers to the accuracy of the initial equilibrium capacity prediction model. When the reliability is greater, the fusion coefficient of the corresponding initial equilibrium capacity prediction model is greater, that is, the proportion of the predicted value obtained by the corresponding initial equilibrium capacity prediction model is greater, so as to achieve the closest and more accurate result and lay the foundation for the setting of the subsequent electrothermal conversion rate control method.

[0113] In some embodiments of this application, generating a future time period balance capacity prediction index based on a balance capacity prediction model includes:

[0114] Acquire real-time characteristic operation data of multi-energy system during preset monitoring period, analyze the real-time characteristic operation data during preset monitoring period, and determine the real-time fluctuation characteristics of real-time characteristic operation data during preset monitoring period.

[0115] All real-time feature operation data and the corresponding real-time fluctuation features of the real-time feature operation data are input into the balance capacity prediction model to obtain the balance capacity prediction index for future periods.

[0116] In this embodiment, the preset monitoring period and the period to be analyzed are of the same length. The real-time fluctuation characteristics include the fluctuation speed, fluctuation value, and fluctuation trend of the real-time feature running data within the preset monitoring period.

[0117] In this embodiment, the balance capacity prediction model analyzes the real-time characteristic operation data of the preset monitoring period to obtain the balance capacity prediction index, the predicted fluctuation value, the predicted fluctuation speed, and the predicted fluctuation trend of the balance capacity prediction index in the future period. This lays the data foundation for the subsequent formulation of a reasonable control scheme for electrothermal conversion rate, improves the effectiveness and accuracy of the control method, and ensures the stable operation of the multi-energy system.

[0118] In some embodiments of this application, an initial control strategy is determined by analyzing the balance capability prediction index and the real-time electrothermal conversion rate based on a preset control analysis model, including:

[0119] The current balance capability prediction index is analyzed based on the preset control analysis model to obtain the preset control strategy library corresponding to the current balance capability prediction index.

[0120] The preset control strategy library includes several preset electrothermal conversion rate differences of the current balance capability prediction index, and each preset electrothermal conversion rate difference is associated with a specific preset control strategy.

[0121] The real-time electrothermal conversion rate is obtained, and the difference between the real-time electrothermal conversion rate and the standard electrothermal conversion rate corresponding to the current balance capacity prediction index is calculated to obtain the real-time electrothermal conversion rate difference.

[0122] A similarity analysis is performed between the real-time electrothermal conversion rate difference and several preset electrothermal conversion rate differences in the preset control strategy library. The preset control strategy corresponding to the preset electrothermal conversion rate difference with the highest similarity is set as the initial control strategy for the real-time electrothermal conversion rate.

[0123] In this embodiment, the preset control analysis model includes multiple historical balance capability indices and the optimal historical control strategy for each historical electrothermal conversion rate under each historical balance capability index. The historical application degree of the optimal historical control strategy is greater than the preset application degree threshold and is the one with the highest historical application degree among all historical control strategies for the corresponding historical electrothermal conversion rate.

[0124] In this embodiment, similarity refers to the degree of similarity between the real-time electrothermal conversion rate difference and the preset electrothermal conversion rate difference. The maximum similarity means that the difference between the real-time electrothermal conversion rate difference and the preset electrothermal conversion rate difference is the smallest.

[0125] In this embodiment, a preset control strategy library for the balance capability prediction index is determined by a preset control analysis model. An initial control strategy is determined based on the similarity between the real-time electrothermal conversion rate difference and the preset electrothermal conversion rate difference. This initial control strategy is used to generate control commands for the real-time electrothermal conversion rate. A verification period is set, and the initial control strategy is continuously optimized through real-time feedback to achieve the best control effect of the electrothermal conversion rate and ensure the stable operation of the multi-energy system.

[0126] In some embodiments of this application, obtaining the actual application degree of the initial control strategy within the testing period includes:

[0127] Multiple application evaluation metrics are pre-defined;

[0128] Obtain real-time operational data within the testing period, and determine the associated operational data for each application evaluation indicator based on the correlation between the application evaluation indicators and the real-time operational data;

[0129] The associated operational data is compared with the standard operational data of the corresponding application evaluation indicators to obtain the first degree of deviation between the associated operational data and the standard operational data. Based on the first degree of deviation, the sub-application degree of the associated operational data is generated.

[0130] The second degree of deviation between the associated operating data and the standard operating data before the initial control strategy is obtained, and the trend and rate of change of the deviation degree from the first degree of deviation of the associated operating data are generated.

[0131] The first compensation coefficient for the sub-application degree of the corresponding associated operational data is generated based on the trend of change; the second compensation coefficient for the sub-application degree of the corresponding associated operational data is generated based on the rate of change.

[0132] The actual application degree of the initial control strategy is generated based on the sub-application degree of the associated operational data, the first compensation coefficient, the second compensation coefficient, and the weight coefficients of the corresponding application evaluation indicators.

[0133] The formula for calculating the actual application degree is:

[0134]

[0135] Where Y represents the actual application degree, f1 is the weight coefficient of the first application evaluation indicator, n1 is the total number of associated operational data for the first application evaluation indicator, and Z1 s1 c11 is the sub-application degree of the s1th associated running data of the first application evaluation index. s1 c21 is the first compensation coefficient for the s1th associated operational data of the first application evaluation index. s1 Zr is the second compensation coefficient for the s1th associated operational data of the first application evaluation index. s1 Let c1r be the sub-application degree of the sr-th associated runtime data for the r-th application evaluation index. sr c2r is the first compensation coefficient for the sr-th associated operational data of the r-th application evaluation index. sr is the second compensation coefficient for the sr-th associated running data of the r-th application evaluation index, and nr is the total number of associated running data of the r-th application evaluation index.

[0136] In this embodiment, the testing duration refers to the pre-set duration for testing the control effect of the initial control strategy. The correlation between the application evaluation index and the real-time operating data is set based on the correlation between historical operating data and the application evaluation index.

[0137] In this embodiment, the evaluation indicators include system stability, power utilization effectiveness, performance efficiency, and energy efficiency. The trend of deviation includes normal trend, abnormal trend, and no trend. When it is a normal trend, the first compensation coefficient is 1; when it is an abnormal trend, the first compensation coefficient is -1; and when it is no trend, the first compensation coefficient is 0.5. When the rate of change is greater, the second compensation coefficient is greater, and vice versa. The value range of the second compensation coefficient is (0.8, 1.2).

[0138] In this embodiment, the actual control effect of the initial control strategy is judged based on the actual application degree of the initial control strategy to determine whether it meets the requirements, and the deficiencies of the initial control strategy are promptly identified and adjusted to ensure the efficient and stable operation of the multi-energy system.

[0139] In some embodiments of this application, determining whether to generate a correction instruction for the initial control strategy based on the actual balance capability index and actual application degree includes:

[0140] Pre-set the exponential difference threshold;

[0141] Obtain the index difference between the actual balance capability index and the predicted balance capability index at the same time point. If the index difference is less than the index difference threshold, do not send a correction instruction. If the index difference is greater than the index difference threshold, send the first-level correction instruction of the initial control strategy.

[0142] Pre-set application threshold;

[0143] If the actual application degree is greater than the application degree threshold, no correction instruction will be sent;

[0144] If the actual application degree is less than the application degree threshold, send a secondary correction instruction for the initial control strategy.

[0145] In this embodiment, the first-level correction instruction analyzes the actual balance capability index based on the preset control analysis model, re-determines the initial control strategy or fine-tunes the current initial control strategy, and adjusts the balance capability prediction model. The second-level correction instruction adjusts the initial control strategy based on the difference between the actual application degree and the application degree threshold.

[0146] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and substitutions can be made without departing from the technical principles of this application, and these improvements and substitutions should also be considered within the scope of protection of this application.

Claims

1. A method for controlling the electrothermal conversion rate based on balance capacity prediction, characterized in that, include: Obtain historical operation logs of multi-energy systems, analyze the historical operation data and historical balance capacity index in each historical operation log, and determine characteristic operation data; A balance capacity prediction model for a multi-energy system is constructed based on characteristic operating data and historical balance capacity indices. A balance capacity prediction index for future periods is then generated based on the balance capacity prediction model. Based on the preset control analysis model, the balance capacity prediction index and real-time electrothermal conversion rate are analyzed to determine the initial control strategy, and control commands for the real-time electrothermal conversion rate are generated based on the initial control strategy. Set the test duration, obtain the actual application degree of the initial control strategy within the test duration, and determine whether to generate a correction instruction for the initial control strategy based on the actual balance capability index and the actual application degree. Based on the analysis of the pre-set control analysis model, the balance capacity prediction index and real-time electrothermal conversion rate are analyzed to determine the initial control strategy, including: The current balance capability prediction index is analyzed based on the preset control analysis model to obtain the preset control strategy library corresponding to the current balance capability prediction index. The preset control strategy library includes several preset electrothermal conversion rate differences of the current balance capability prediction index, and each preset electrothermal conversion rate difference is associated with a specific preset control strategy. The real-time electrothermal conversion rate is obtained, and the difference between the real-time electrothermal conversion rate and the standard electrothermal conversion rate corresponding to the current balance capacity prediction index is calculated to obtain the real-time electrothermal conversion rate difference. A similarity analysis is performed between the real-time electrothermal conversion rate difference and several preset electrothermal conversion rate differences in the preset control strategy library. The preset control strategy corresponding to the preset electrothermal conversion rate difference with the highest similarity is set as the initial control strategy for the real-time electrothermal conversion rate. The actual application degree of the initial control strategy within the test period is obtained, including: Multiple application evaluation metrics are pre-defined; Obtain real-time operational data within the testing period, and determine the associated operational data for each application evaluation indicator based on the correlation between the application evaluation indicators and the real-time operational data; The associated operational data is compared with the standard operational data of the corresponding application evaluation indicators to obtain the first degree of deviation between the associated operational data and the standard operational data. Based on the first degree of deviation, the sub-application degree of the associated operational data is generated. The second degree of deviation between the associated operating data and the standard operating data before the initial control strategy is obtained, and the trend and rate of change of the deviation degree from the first degree of deviation of the associated operating data are generated. The first compensation coefficient for the sub-application degree of the corresponding associated operational data is generated based on the changing trend. The second compensation coefficient is generated based on the rate of change, corresponding to the sub-application degree of the associated operational data; based on the sub-application degree of the associated operational data, the first compensation coefficient, the second compensation coefficient, and... The weight coefficients of the corresponding application evaluation indicators are used to generate the actual application degree of the initial control strategy; The formula for calculating the actual application degree is: Practical application degree, f1 is the weight coefficient of the first application evaluation indicator, n1 is the total number of associated operational data of the first application evaluation indicator, Z1 s1 c11 is the sub-application degree of the s1th associated running data of the first application evaluation index. s1 c21 is the first compensation coefficient for the s1th associated operational data of the first application evaluation index. s1 Zr is the second compensation coefficient Zr for the s1th associated operational data of the first application evaluation index. s1 Let c1r be the sub-application degree of the sr-th associated runtime data for the r-th application evaluation index. sr c2r is the first compensation coefficient for the sr-th associated operational data of the r-th application evaluation index. sr is the second compensation coefficient for the sr-th associated running data of the r-th application evaluation index, and nr is the total number of associated running data of the r-th application evaluation index.

2. The electrothermal conversion rate control method based on balance capability prediction as described in claim 1, characterized in that, Analyze the historical operation data and historical balancing capacity index in each historical operation log to identify characteristic operation data, including: Acquire historical operation logs of multiple historical monitoring cycles of the multi-energy system, and determine the historical balance capacity index at each acquisition time node based on a preset balance capacity index algorithm for each historical operation log. Using the historical monitoring duration of each historical monitoring cycle as the time reference line and the preset time interval as the collection time node, the historical operation data and historical balance capacity index in the historical operation log of the corresponding historical monitoring cycle are obtained according to the collection time node and mapped to the corresponding time reference line to obtain the impact analysis diagram of each historical monitoring cycle. For each impact analysis diagram, the historical balance capability index of all collection time nodes is analyzed for fluctuation. Collection time nodes with fluctuation of historical balance capability index greater than the preset fluctuation threshold are selected. The selected adjacent collection time nodes are set as a characteristic fluctuation period, and the period to be analyzed for each characteristic fluctuation period is determined. Obtain historical operational data for each time period to be analyzed in the same impact analysis diagram, and determine whether the degree of change of each historical operational data for the time period to be analyzed is greater than the preset degree of change threshold. If not, remove the corresponding historical operational data; if so, retain the corresponding historical operational data. Obtain the first fluctuation feature of the retained historical operating data for each period to be analyzed and the second fluctuation feature of the historical balance capability index for the corresponding period of characteristic fluctuation. Determine the characteristic operating data based on the first and second fluctuation features.

3. The electrothermal conversion rate control method based on balance capability prediction as described in claim 2, characterized in that, Based on the first and second fluctuation characteristics, characteristic operation data is determined, including: Determine the initial mutation time node of the retained historical running data in the characteristic period to be analyzed, and obtain the first fluctuation feature of each retained historical running data in the initial mutation time node and the characteristic period to be analyzed. The first fluctuation feature includes the fluctuation value, fluctuation speed and fluctuation trend of the corresponding historical running data at multiple preset analysis time nodes in the initial mutation time node and the characteristic period to be analyzed. Obtain the second fluctuation feature of the historical balance capability index for each characteristic fluctuation period corresponding to the characteristic fluctuation period to be analyzed. The second fluctuation feature includes the fluctuation value, fluctuation rate and fluctuation trend of the historical balance capability index at multiple preset analysis time nodes in the corresponding characteristic fluctuation period. The second fluctuation feature of the historical balance capability index during the characteristic fluctuation period is compared with the first fluctuation feature of each historical operating data of the period to be analyzed corresponding to the characteristic fluctuation period. Historical operating data with a similarity greater than the preset similarity threshold are selected and set as the operating data of interest for the corresponding characteristic fluctuation period. Pre-set a preset balance capability index range to determine the preset balance capability index range in which the historical balance capability index of all characteristic fluctuation periods in different influence analysis charts falls; By comparing the operational data of each characteristic fluctuation period within the same preset balance capability index range, the frequency of occurrence of the same operational data of each characteristic fluctuation period within the same preset balance capability index range can be obtained. If the frequency of occurrence is less than the preset frequency threshold, the corresponding data of interest is removed. If the frequency of occurrence is greater than the preset frequency threshold, the corresponding data of interest is retained and the retained data of interest is set as feature data.

4. The electrothermal conversion rate control method based on balance capability prediction as described in claim 3, characterized in that, The fluctuation trend of each historical operating data in the period to be analyzed at all preset analysis time nodes is compared with the fluctuation trend of the historical balance capability index of the characteristic fluctuation period at all preset analysis time nodes to obtain the degree of deviation between the fluctuation trend of each historical operating data and the historical balance capability index. Based on the principle of time node comparison, each historical operating data in the period to be analyzed is matched with the preset analysis time node of the historical balance ability index of the characteristic fluctuation period to obtain multiple preset analysis time node groups. Obtain the fluctuation values ​​of the first fluctuation feature and the second fluctuation feature of each preset analysis time node group and calculate the difference to obtain the fluctuation value difference of each preset analysis time node group; The fluctuation velocities of the first and second fluctuation characteristics of each preset analysis time node group are obtained and subtracted to obtain the fluctuation velocity difference value of each preset analysis time node group; The similarity between each historical operating data point and the corresponding historical balance capability index is generated based on the degree of deviation in the fluctuation trend, the difference in the fluctuation value of each preset analysis time node group, and the difference in the fluctuation speed. The formula for calculating the similarity is: Where D is the similarity, x1 is the first similarity conversion coefficient, q1 is the weight coefficient of the fluctuation trend, b1 is the degree of deviation of the fluctuation trend, x2 is the second similarity conversion coefficient, q2 is the weight coefficient of the fluctuation value, b2i is the difference of the fluctuation value of the i-th preset analysis time node group, m is the total number of preset analysis time node groups, x3 is the third similarity conversion coefficient, q3 is the weight coefficient of the fluctuation speed, and b3i is the difference of the fluctuation speed of the i-th preset analysis time node group.

5. The electrothermal conversion rate control method based on balance capability prediction as described in claim 4, characterized in that, A multi-energy system balance capacity prediction model is constructed based on characteristic operating data and historical balance capacity indices, including: Obtain the second fluctuation characteristics of the historical balance capacity index for all characteristic fluctuation periods, and the first fluctuation characteristics of the characteristic operation data of the period to be analyzed corresponding to the characteristic fluctuation periods; Based on the same characteristic running data and the first fluctuation feature of the same characteristic running data for all the periods to be analyzed, the data is divided into training input data and test input data. The historical balance ability index of the characteristic fluctuation period corresponding to the period to be analyzed in the training input data and the second fluctuation feature of the historical balance ability index are used as training output data to obtain the initial balance ability prediction model of each characteristic running data and the historical balance ability index. The credibility of the corresponding initial balance capability prediction model is generated based on the test input data. The initial balance capability prediction model with a credibility of less than the preset credibility threshold is iteratively trained until the credibility is greater than the preset credibility threshold. The fusion coefficient of each initial equilibrium capacity prediction model is set according to the credibility of each initial equilibrium capacity prediction model; All initial balance capacity prediction models and their corresponding fusion coefficients are fused, and the fused model is set as the balance capacity prediction model for a multi-energy system.

6. The electrothermal conversion rate control method based on balance capability prediction as described in claim 5, characterized in that, The equilibrium capacity prediction model generates an equilibrium capacity prediction index for future periods, including: Acquire real-time characteristic operation data of multi-energy system during preset monitoring period, analyze the real-time characteristic operation data during preset monitoring period, and determine the real-time fluctuation characteristics of real-time characteristic operation data during preset monitoring period. All real-time feature operation data and the corresponding real-time fluctuation features of the real-time feature operation data are input into the balance capacity prediction model to obtain the balance capacity prediction index for future periods.

7. The electrothermal conversion rate control method based on balance capability prediction as described in claim 6, characterized in that, Determine whether to generate correction instructions for the initial control strategy based on the actual balance capability index and actual application degree, including: Pre-set the exponential difference threshold; Obtain the index difference between the actual balance capability index and the predicted balance capability index at the same time point. If the index difference is less than the index difference threshold, do not send a correction instruction. If the index difference is greater than the index difference threshold, send the first-level correction instruction of the initial control strategy. Pre-set application threshold; If the actual application degree is greater than the application degree threshold, no correction instruction will be sent; If the actual application degree is less than the application degree threshold, send a secondary correction instruction for the initial control strategy.

Images