A heat supply network scheduling method and system based on an agent

Abstract

This invention provides a heating network scheduling method and system based on intelligent agents, belonging to the field of data processing technology. Specifically, it includes: using historical scheduling data of the heating network, determining joint scheduling processing data for different state-changing networks; determining the scheduling impact of heating networks between heat stations of different state-changing networks in different joint scheduling processing cycles; determining the correlation coefficients of different state-changing networks based on the scheduling impact; determining the correlation of changing network groups composed of different state-changing networks based on the correlation coefficients of different state-changing networks; and determining a characteristic parameter identification and correction method for the hydraulic simulation model based on the correlation, thereby improving the reliability of load scheduling processing.

Description

Technical Field

[0001] This invention belongs to the field of intelligent agent technology, and in particular relates to a method and system for scheduling heating pipeline networks based on intelligent agents. Background Technology

[0002] In existing technical solutions, load scheduling for heating networks is often based on load scheduling instructions from operators. The specific load scheduling strategy is determined according to the current load gap or hydraulic balance risk. However, all of the above technical solutions suffer from the following technical problems: To address the aforementioned technical problems, this application provides a heating network scheduling method and system based on intelligent agents. Summary of the Invention

[0003] To achieve the objectives of this invention, the following technical solution is adopted: Specifically, this application provides a heating network scheduling method based on intelligent agents, which includes: S1 uses load regulation data to determine load regulation anomaly data under different weather types, and based on the load regulation anomaly data, determines the load regulation anomaly type and associated abnormal heat stations under different weather types. By utilizing the degree of overlap of associated abnormal heat stations under different load regulation anomaly types, the update setting strategy of the heat station's intelligent agent is determined. S2 uses the agent's update setting strategy to determine the agent's setting data, and combines the agent's setting data with the overlap of load adjustment heat stations under different weather types to determine the response optimization weather type in the weather type. S3 uses the identification and matching data of agents regarding load adjustment needs in different response-optimized weather types, and combines this with the updated matching results of the agents to determine the update strategy for the response-optimized weather type in the aforementioned weather type.

[0004] Furthermore, the abnormal load regulation data is determined based on whether the heating supply after load scheduling processing under the weather type can meet the heating demand.

[0005] Furthermore, the method for determining the type of load regulation anomaly under the aforementioned weather type is as follows: S11 uses the load regulation anomaly data under the aforementioned weather type to determine the number of times the heating station's heating supply cannot meet the heating demand after load scheduling processing, and uses this as the number of regulation deviations. S12 uses the number of adjustment deviations to determine the associated abnormal heat stations under the weather type; S13 determines the load regulation anomaly type under the weather type based on the associated abnormal heat station data under the weather type.

[0006] Furthermore, the method for determining the response-optimized weather type in the aforementioned weather type is as follows: S31 uses the setting data of the intelligent agent to determine the heat station with the intelligent agent set, and uses it as the set heat station; S32 determines the degree of overlap between the set-up heating station and the associated heating station under the weather type based on the overlap between the set-up heating station and the heating station for load regulation under different weather types. S33 determines the response-optimized weather type in the weather type based on the setting of the heat station and the degree of overlap between the set heat station and the associated heat station under the weather type.

[0007] On the other hand, this application provides a computer system that stores a computer program. When the computer program is executed in the computer, it causes the computer to execute the above-mentioned intelligent agent-based heating network scheduling method. Attached Figure Description

[0008] The above and other features and advantages of the present invention will become more apparent from a detailed description of exemplary embodiments thereof with reference to the accompanying drawings.

[0009] Figure 1 This is a flowchart of a heating network scheduling method based on intelligent agents; Figure 2 This is a flowchart illustrating the method for determining the type of load regulation anomaly under different weather conditions; Figure 3 This is a flowchart illustrating the method for determining the update settings strategy of the intelligent agent in a heating station. Detailed Implementation

[0010] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many ways and should not be construed as limited to the embodiments set forth herein; rather, they are provided so that the invention will be thorough and complete, and the concept of the exemplary embodiments will be fully conveyed to those skilled in the art. The same reference numerals in the drawings denote the same or similar structures, and therefore their detailed description will be omitted.

[0011] The terms “a,” “one,” “the,” and “the” are used to indicate the existence of one or more elements / components / etc.; the terms “including” and “having” are used to indicate an open-ended meaning of inclusion and that other elements / components / etc. may exist in addition to the listed elements / components / etc.

[0012] Example 1 To solve the above problems, according to one aspect of the present invention, such as Figure 1 As shown, a heating network scheduling method based on intelligent agents is provided, specifically including: S1 uses load regulation data to determine load regulation anomaly data under different weather types, and based on the load regulation anomaly data, determines the load regulation anomaly type and associated abnormal heat stations under different weather types. By utilizing the degree of overlap of associated abnormal heat stations under different load regulation anomaly types, the update setting strategy of the heat station's intelligent agent is determined. S2 uses the agent's update setting strategy to determine the agent's setting data, and combines the agent's setting data with the overlap of load adjustment heat stations under different weather types to determine the response optimization weather type in the weather type. S3 uses the identification and matching data of agents regarding load adjustment needs in different response-optimized weather types, and combines this with the updated matching results of the agents to determine the update strategy for the response-optimized weather type in the aforementioned weather type.

[0013] Furthermore, the abnormal load regulation data is determined based on whether the heating supply after load scheduling processing under the weather type can meet the heating demand.

[0014] Specifically, such as Figure 2 As shown, the method for determining the load regulation anomaly type under the aforementioned weather type is as follows: In this embodiment, based on the load regulation anomaly data under the weather type, the distribution data of heat stations with poor heating regulation reliability under the weather type are determined, and the load regulation anomaly type under the weather type is determined based on the heat station data with poor heating regulation reliability. This also lays the foundation for determining the construction strategy of intelligent agents in the heat station in a targeted manner.

[0015] S11 uses the load regulation anomaly data under the aforementioned weather type to determine the number of times the heating station's heating supply cannot meet the heating demand after load scheduling processing, and uses this as the number of regulation deviations. The weather type refers to the weather classification result formed by classifying dates with temperature, humidity, and wind speed in the same range into the same category; the load regulation anomaly data refers to historical data where the heating supply fails to meet the heating demand after load scheduling under the weather type; the number of regulation deviations refers to the cumulative number of times a heating station fails to meet the heating demand after load scheduling under a certain weather type.

[0016] Assuming the heating network covers multiple heating stations, and various weather types are obtained by dividing the ranges of temperature, humidity, and wind speed, the number of times each heating station failed to meet the heating demand during its historical operation under each weather type is counted as the number of adjustment deviations of that heating station under that weather type.

[0017] This step quantifies and statistically analyzes the abnormal regulation situations of each heating station under different weather conditions. Its significance lies in the fact that by accurately counting the number of regulation deviations, the reliability of load regulation of each heating station under different weather conditions can be objectively reflected, avoiding incorrect setting decisions based solely on subjective judgment or a single abnormal event. This provides a reliable data foundation for the subsequent identification of associated abnormal heating stations.

[0018] S12 uses the number of adjustment deviations to determine the associated abnormal heat stations under the weather type; The associated abnormal heating stations refer to heating stations whose daily average adjustment deviation number is greater than a preset deviation number threshold under the weather type; the preset deviation number threshold refers to a pre-set daily average limit for judging whether the degree of adjustment abnormality of a heating station under a certain weather type has reached the attention level.

[0019] Assuming there are multiple heating stations in the heating network, calculate the average daily adjustment deviation number for each heating station under a certain weather type. For heating stations whose average daily adjustment deviation number exceeds the preset deviation number threshold, mark them as associated abnormal heating stations under that weather type.

[0020] This step filters each heating station by setting a daily average deviation threshold. Its significance lies in identifying heating stations with more prominent regulation anomalies from all heating stations, focusing on heating stations that truly have regulation reliability issues, and avoiding including heating stations with occasional regulation deviations in the anomaly category. This improves the targeting and accuracy of subsequent anomaly type judgment and agent setting strategy determination.

[0021] S13 determines the load regulation anomaly type under the weather type based on the associated abnormal heat station data under the weather type.

[0022] The load regulation anomaly type refers to the degree of anomaly determined according to the number of associated abnormal heat stations under the weather type, including Class I, Class II and Class III anomaly types. Among them, the load regulation anomaly degree of Class I anomaly type is greater than that of Class II anomaly type, and the load regulation anomaly degree of Class II anomaly type is greater than that of Class III anomaly type. The corresponding preset anomaly type refers to the corresponding anomaly type mapping relationship pre-set according to the range of the number of associated abnormal heat stations.

[0023] If a certain weather type has a large number of associated abnormal heat stations, it is classified as a type 1 abnormality; if the number of associated abnormal heat stations is moderate, it is classified as a type 2 abnormality; if the number of associated abnormal heat stations is small, it is classified as a type 3 abnormality, thus completing the determination of the load regulation abnormality type under that weather type.

[0024] This step maps the number of associated abnormal heating stations to the corresponding abnormality type level, realizing the quantitative classification of the degree of abnormality in the weather type dimension. Its significance lies in the fact that different load regulation abnormality types directly reflect the overall regulation pressure faced by the heating network under that weather type. This provides a standardized classification basis for the subsequent statistical analysis of the number of abnormal weather types and the decision-making of intelligent agents in S21, thereby making the formulation of scheduling strategies more targeted and operable.

[0025] This embodiment achieves a complete transformation from raw load regulation anomaly data to weather-type load regulation anomaly types through steps S11 to S13. Its core value lies in three aspects: First, it objectively identifies the regulation anomaly patterns of each heating station under different weather types through quantitative statistics of the number of regulation deviations; second, it filters out the truly relevant and abnormal heating stations that need attention through the daily average deviation threshold, avoiding the influence of interfering information; and third, it provides standardized input indicators for determining the subsequent intelligent agent setting strategy through the hierarchical mapping of anomaly types.

[0026] In this specific embodiment, the heating network covers 20 heating stations, and the historical operation data covers 3 heating seasons. According to the combination of three dimensions such as temperature (low temperature range: <-5℃, medium temperature range: -5℃ to 5℃, high temperature range: >5℃), a total of 6 effective weather types were obtained, which are denoted as W1 to W6 respectively.

[0027] S11 execution process: The number of times each heating station failed to meet heating demand during its historical operation under each weather type was recorded. Taking heating station H1 as an example: H1 experienced a total of 32 adjustment deviations under weather type W1. The historical corresponding number of days for weather type W1 is 40 days. The average number of adjustment deviations of H1 under W1 is 32 ÷ 40 = 0.80 times / day.

[0028] S12 execution process: The preset threshold for the number of deviations was set to 0.50 times / day. The average daily adjustment deviation was calculated for each of the 20 heating stations under each weather type. Heating stations with an average daily deviation greater than 0.50 times / day were selected as associated abnormal heating stations for that weather type. The results are as follows: There are 8 associated anomalous thermal stations under weather type W1 (H1, H3, H5, H7, H9, H12, H15, H18); 5 associated anomalous thermal stations under weather type W2 (H2, H5, H7, H11, H16); 3 associated anomalous thermal stations under weather type W3 (H4, H8, H14); and 2, 1, and 0 associated anomalous thermal stations under weather types W4, W5, and W6, respectively.

[0029] S13 execution process: The preset anomaly type mapping relationship is as follows: ≥6 associated anomaly heat stations correspond to type 1 anomaly; 3 ≤ number < 6 correspond to type 2 anomaly; 0 < number < 3 correspond to type 3 anomaly; and 0 corresponds to no anomaly type.

[0030] Based on this, the load regulation anomaly types for each weather type are determined as follows: W1 corresponds to Class I anomaly type (8); W2 corresponds to Class II anomaly type (5); W3 corresponds to Class II anomaly type (3); W4 corresponds to Class III anomaly type (2); W5 corresponds to Class III anomaly type (1); and W6 corresponds to Class III anomaly type (0).

[0031] Specifically, such as Figure 3 As shown, the method for determining the update setting strategy of the intelligent agent in the heating station is as follows: In this embodiment, based on the degree of abnormality and overlap of the load regulation of heating stations under different weather types, a comprehensive assessment of the overall heating deviation risk of the heating network is conducted. Using the heating deviation risk and the heating deviation risk of the heating stations, a targeted update setting strategy for the intelligent agents of the heating stations is determined, thereby laying the foundation for improving the regulation reliability of the heating network.

[0032] S21 determines the number of weather types under different load adjustment anomaly types based on the load adjustment anomaly type of the weather type; The number of weather types under different load regulation anomaly types refers to the number of weather types that are classified as Class I, Class II, and Class III anomaly types among all valid weather types.

[0033] Assuming that there are multiple valid weather types in the heating network, the number of weather types identified as Category I, Category II, and Category III anomalous weather types will be counted separately as the basis for subsequent situation judgment.

[0034] This step completes the statistical analysis of the distribution scale of each type of anomaly. Its significance lies in the fact that the number of a type of weather anomaly directly reflects the coverage of the high-intensity regulation pressure faced by the heating network. The more the number, the more severe the challenges to the overall regulation reliability under various weather conditions. Therefore, using the number of a type of weather anomaly as the first judgment criterion can quickly identify the overall high-risk state of the heating network, thereby providing guidance for the determination of subsequent targeted setting strategies.

[0035] It should be noted that, in case 1: if the number of weather types under a certain abnormal type is greater than the preset threshold for the number of weather types, then if the heating station has an adjustment deviation in the weather types above the second preset number, then the intelligent agent setting process will be performed. The preset weather type quantity threshold refers to a pre-set limit value used to determine whether the quantity of a certain type of abnormal weather has reached a high overall risk level; the second preset quantity refers to the lower limit of the number of adjustment deviation weather types used to determine whether a single heating station needs to be processed by intelligent agent settings under case 1.

[0036] If the number of weather types under a certain abnormality exceeds the preset threshold, it indicates that the heating network as a whole is in a high-risk state. Then, each heating station is checked one by one. If a heating station has adjustment deviations in all weather types above the second preset number, then the intelligent agent settings are applied to that heating station.

[0037] This situation directly triggers the intelligent agent setting decision of the heating station when the coverage of a certain type of abnormal weather is wide. Its significance lies in the fact that when the heating network as a whole faces large-scale high-intensity regulation pressure, the problem of regulation reliability has become systematic. Prioritizing the intelligent agent setting of heating stations with regulation deviations across weather types can quickly improve the overall coordinated regulation capability of the heating network under high-risk weather conditions, thereby effectively reducing the risk of large-scale heating failure.

[0038] Additionally, it should be noted that in S21, Case 2: if the number of weather types under a certain type of abnormality is not greater than the preset threshold for the number of weather types, it is determined whether there is a weather type with a certain type of abnormality among the weather types. If yes, proceed to step S22. If no, when the heating station has an adjustment deviation in the weather types above the preset number, and there is a weather type with a second type of abnormality or a weather type with a certain type of abnormality among the weather types with adjustment deviation, the setting process of the intelligent agent is performed. The preset quantity refers to the lower limit of the number of weather types with adjustment deviations used to determine whether a single heating station needs to be processed by intelligent agent settings when there is no type of abnormality in case 2, and the preset quantity is greater than the second preset quantity; the presence of two types of abnormality or one type of abnormality among the weather types with adjustment deviations means that at least one of the weather types in which the heating station has adjustment deviations belongs to two types of abnormality or one type of abnormality, indicating that the adjustment deviation of the heating station does not only occur in low-risk weather types.

[0039] Assuming that the number of weather types of a certain type of anomalous weather does not exceed the preset threshold for the number of weather types, and there is no anomalous weather type among the weather types, then each heating station is checked one by one: if a heating station has an adjustment deviation in weather types above the preset number, and at least one of the weather types belongs to either a second-type or a first-type anomalous weather type, then the intelligent agent settings are performed for that heating station.

[0040] In this situation, when the coverage area of ​​a certain type of abnormal weather is small and there is no such type of abnormal weather, the setting decision is made by comprehensively considering the number of weather types covered by the regulation deviation of the heating station and the severity of the deviation weather types. The significance is to avoid ignoring the local regulation risks of individual heating stations under medium and high intensity weather conditions because the overall risk is not high, thereby ensuring the accuracy and coverage integrity of the setting strategy.

[0041] S22 determines the weather type of the heat station belonging to the associated abnormal heat station based on the degree of overlap of the associated abnormal heat stations under different load regulation abnormal weather types, and uses it as the associated abnormal weather type of the heat station. The associated abnormal weather type of the heat station refers to the set of weather types corresponding to which a heat station is identified as an associated abnormal heat station in different weather types; the degree of overlap refers to the degree to which the same heat station is identified as an associated abnormal heat station in multiple weather types.

[0042] Suppose that a certain heating station in the heating network is identified as an associated abnormal heating station under three weather types: W1, W2, and W3. Then W1, W2, and W3 are the associated abnormal weather types of this heating station, and the number of associated abnormal weather types of this heating station is 3.

[0043] This step, by summarizing the abnormal situations of various heating stations across different weather types, identifies heating stations that consistently have regulation reliability issues under multiple weather types. The significance of this step is that heating stations with a large number of associated abnormal weather types indicate that their regulation anomalies are not isolated incidents affected by specific weather conditions, but rather have strong universality. Therefore, setting up intelligent agents for these heating stations has higher priority and more significant improvement value, thereby concentrating intelligent agent setting resources on the heating stations that most need improvement.

[0044] It also includes the following: determining whether there are heat stations with a number of associated abnormal weather types greater than a preset threshold for the number of associated abnormal weather types; if so, if the heat station has an adjustment deviation in weather types greater than a second preset number, then the intelligent agent setting process is performed; otherwise, proceed to step S23. The preset threshold for the number of associated abnormal weather types refers to a pre-set limit value used to determine whether the number of associated abnormal weather types at a certain heating station has reached the level that requires priority attention.

[0045] If some heating stations have more than the preset threshold for the number of associated abnormal weather types, then these heating stations are checked one by one: if there is an adjustment deviation in the weather types above the second preset number, then the intelligent agent is set up for them; if the number of associated abnormal weather types of all heating stations does not exceed the preset threshold, then the process proceeds to step S23 for further judgment.

[0046] This judgment logic filters out heat stations with prominent cross-weather anomalies by associating abnormal weather types with a threshold and prioritizes their setting decisions. Its significance lies in timely identifying high-priority heat stations that have a continuous risk of regulation under multiple weather conditions, avoiding the omission of such heat stations in subsequent steps and causing processing delays, thereby ensuring the timeliness and effectiveness of the agent's setting work.

[0047] S23 uses the number of weather types under different load adjustment anomaly types and the associated abnormal weather types of different heating stations to determine the update setting strategy of the intelligent agent of the heating station. Furthermore, based on the proportion of heating stations with associated abnormal weather types, the load regulation deviation coefficient of the heating network is determined. It is then determined whether the load regulation deviation coefficient of the heating network is greater than a preset deviation coefficient threshold. If so, if the heating station has a regulation deviation in more than a second preset number of weather types, the intelligent agent is set up. If not, the intelligent agent is set up when the heating station has a regulation deviation in more than a preset number of weather types. The load regulation deviation coefficient of the heating network refers to a comprehensive assessment coefficient reflecting the overall regulation deviation risk, determined based on the proportion of heating stations with associated abnormal weather types to all heating stations in the heating network; the preset deviation coefficient threshold refers to a pre-set threshold value used to determine whether the overall load regulation deviation risk of the heating network has reached a high-risk level.

[0048] If the heating network has multiple heating stations, and the proportion of heating stations with associated abnormal weather types is relatively high, then the calculated load regulation deviation coefficient will be large, exceeding the preset deviation coefficient threshold. Therefore, a lower threshold for the number of weather types with regulation deviation will be used for agent setting. If the proportion is low and the deviation coefficient does not exceed the threshold, then a higher threshold will be used for setting.

[0049] This step determines the final intelligent agent setting strategy for the remaining heating stations by using the overall load adjustment deviation coefficient of the heating network. Its significance lies in the fact that after the high-risk heating stations have been dealt with in the previous steps, the setting decision for the remaining heating stations needs to be dynamically adjusted in combination with the overall risk level of the heating network. When the overall risk is high, the setting threshold for a single station should be appropriately lowered to expand the coverage. When the overall risk is low, the setting threshold should be raised to avoid over-setting and wasting resources, thereby achieving the optimal allocation of intelligent agent setting resources under the overall risk assessment framework.

[0050] The preset quantity is greater than the second preset quantity. That is, when the overall deviation coefficient of the heating network is high, a lower threshold for the number of weather types for adjusting deviation of a single station is used for setting, and when the overall deviation coefficient is low, a higher threshold is used for setting.

[0051] Following the specific implementation of S1, the abnormal load regulation types for each weather type have been determined: W1 is a Class I abnormal type, W2 and W3 are Class II abnormal types, W4, W5 and W6 are Class III abnormal types, and W6 has no abnormal type.

[0052] S21 execution process: The number of weather types under each anomaly type is counted as follows: the number of weather types under the first type of anomaly is 1 (W1); the number of weather types under the second type of anomaly is 2 (W2, W3); and the number of weather types under the third type of anomaly is 3 (W4, W5, W6).

[0053] The preset threshold for the number of weather types is set to 2. The number of one type of abnormal weather is 1, which is not greater than the preset threshold of 2, so the case proceeds to condition 2.

[0054] Case 2: There is an abnormal type (W1) in the weather type, so proceed to step S22.

[0055] S22 execution process: The associated abnormal weather types for each heat station were statistically analyzed: H1 was an associated abnormal heat station under both weather types W1 and W2, with a total of 2 associated abnormal weather types; H5 was an associated abnormal heat station under both weather types W1 and W2, with a total of 2 associated abnormal weather types; H7 was an associated abnormal heat station under both weather types W1 and W2, with a total of 2 associated abnormal weather types; the number of associated abnormal weather types for the remaining heat stations did not exceed 1.

[0056] The preset threshold for the number of associated abnormal weather types is set to 2. There are heat stations (H1, H5, H7) with a number of associated abnormal weather types greater than or equal to 2. Here, it is judged that "greater than the preset threshold" is true. The threshold is 2. The number of H1, H5, and H7 is equal to 2, which is not greater than 2. Therefore, there are no heat stations with a number greater than the preset threshold. Proceed to step S23.

[0057] S23 execution process: Let A be the number of heating stations with associated abnormal weather types, and B be the total number of heating stations in the heating network. Then: Load adjustment deviation coefficient = A ÷ B. In this embodiment, the heating stations with associated abnormal weather types are H1, H3, H5, H7, H9, H12, H15, H18 (from W1) and H2, H11, H16 (from W2 and not appearing in W1), totaling 11.

[0058] Substituting the specific values ​​into the formula: Load adjustment deviation coefficient = 11 ÷ 20 = 0.55; The preset deviation coefficient threshold is set to 0.50. Since the load regulation deviation coefficient of 0.55 is greater than the preset deviation coefficient threshold of 0.50, a second preset quantity threshold (the second preset quantity is set to 2) is used to judge each heating station: if a heating station has regulation deviation in 2 or more weather types, then the intelligent agent is set for it.

[0059] After verification, H1, H5, and H7 all showed adjustment deviations in two or more weather types. Therefore, intelligent agent settings were applied to H1, H5, and H7. The remaining heating stations showed adjustment deviations only in one or zero weather types, and no adjustments were made.

[0060] Final agent configuration data: The number of heat stations with agents is set to 3 (H1, H5, H7), that is, the heat stations are set to H1, H5, and H7.

[0061] Furthermore, the configuration data of the intelligent agent includes the number of heat stations with intelligent agents configured.

[0062] Furthermore, the method for determining the response-optimized weather type in the aforementioned weather type is as follows: In this application, based on the overlap of the heat stations set up by the intelligent agents and the heat stations for load regulation under the aforementioned weather types, the risks of using intelligent agents for collaborative heating regulation under different weather types are determined. Based on the assessment and analysis results of the risks of using intelligent agents for collaborative heating regulation, the response optimization weather type is determined. While ensuring the reliability of heating, the risks of using intelligent agents for collaborative scheduling are also reduced, thus ensuring the reliability of heating.

[0063] S31 uses the setting data of the intelligent agent to determine the heat station with the intelligent agent set, and uses it as the set heat station; The configuration data of the agent refers to the number of heat stations and the specific list of heat stations that need to be configured in S2; the configuration heat station refers to the set of heat stations that need to be configured in the agent update configuration strategy determination process in S2.

[0064] Assuming that after the judgment process in S2, a portion of the heating stations in the heating network are identified as requiring intelligent agent settings, these heating stations are aggregated into a set of set heating stations, which serves as the basic data input for the overlap analysis in S32 and S33.

[0065] This step completes the summary and confirmation of the agent setting results. Its significance lies in unifying and integrating the agent-set heat stations determined in multiple judgment branches in S2 into a clear set of set heat stations. This provides a clear and complete data foundation for subsequent analysis based on the degree of overlap between set heat stations and weather type-related heat stations, and avoids affecting the accuracy of subsequent judgment results due to omissions or duplicate statistics.

[0066] S32 determines the degree of overlap between the set-up heating station and the associated heating station under the weather type based on the overlap between the set-up heating station and the heating station for load regulation under different weather types. The heat station for load adjustment under the weather type refers to the heat station whose average daily load adjustment frequency is greater than the preset adjustment frequency threshold under the weather type, i.e., the associated heat station; the associated heat station refers to the heat station that frequently performs load adjustment under the weather type; the degree of overlap between the set heat station and the associated heat station refers to the size of the intersection between the set heat station and the set of associated heat stations for the weather type, which is usually quantified by the proportion of the set heat station in the associated heat station (i.e., the overlap setting ratio).

[0067] Assuming there are multiple associated heating stations under a certain weather type, some of which are already designated heating stations, the proportion of designated heating stations among the associated heating stations for that weather type is calculated as an indicator of the degree of overlap between designated heating stations and associated heating stations under that weather type.

[0068] This step provides a core basis for judging whether each weather type is suitable as a response optimization weather type by quantifying the coverage ratio of the heating station among the associated heating stations for each weather type. Its significance lies in the fact that only when the heating station occupies a sufficient proportion among the associated heating stations for a certain weather type can the main adjustment needs under that weather type be effectively covered by the use of intelligent agents for coordinated adjustment. Otherwise, even if intelligent agents are used for coordinated scheduling, the reliability of heating cannot be significantly improved. Therefore, the degree of overlap is a key quantitative indicator for judging the response optimization weather type.

[0069] S33 determines the response-optimized weather type in the weather type based on the setting of the heat station and the degree of overlap between the set heat station and the associated heat station under the weather type.

[0070] The optimized weather type refers to a weather type under which intelligent agents can effectively coordinate heating regulation and the regulation risk is controllable; the preset threshold for the number of heating stations refers to a pre-set limit value used to determine whether the number of heating stations meets the minimum reliability requirements for coordinated regulation; the preset matching ratio threshold refers to a pre-set limit value used to determine whether the coverage ratio of the heating stations among all heating stations meets the overall participation requirements; the preset setting ratio threshold refers to a pre-set limit value used to determine whether the overlapping setting ratio meets the sufficiency requirements for coordinated regulation of the weather type.

[0071] If the number of heat stations set does not reach the preset threshold, it means that the current scale of heat stations set by the agent is too small and the reliability of coordinated regulation is poor. Therefore, it is determined that all weather types do not belong to the response optimization weather types.

[0072] This step, through multi-level quantitative and proportional condition judgments, achieves a comprehensive assessment of the suitability of each weather type as a response optimization weather type. Its significance lies in avoiding blindly carrying out response optimization scheduling when the scale of the intelligent agent is insufficient or the overlap of the heating stations associated with the weather type is insufficient. This ensures that response optimization is only initiated under weather types where the collaborative regulation of the intelligent agent can truly play a role, thus guaranteeing the safety and effectiveness of heating collaborative scheduling.

[0073] It should be noted that if the number of heating stations set is less than the preset threshold for the number of heating stations, the reliability of load regulation when combined with the setting of heating stations is poor. Therefore, it is determined that all weather types do not belong to the response optimization weather types. The poor reliability of regulation refers to the fact that when the number of heat stations equipped with intelligent agents is too small, the coverage of the intelligent agents' collaborative regulation is insufficient to form an effective overall collaborative effect. In this case, rashly listing certain weather types as response optimization weather types may cause some heat stations to experience load imbalance.

[0074] If the number of heating stations in the heating network is only 1, which is significantly less than the preset threshold for the number of heating stations, then it is directly determined that all weather types do not belong to the response optimization weather type, and no further judgment is made.

[0075] This situation is the first threshold of the entire S33. Its significance lies in setting a minimum scale requirement to protect the subsequent response optimization weather type determination process, avoiding ineffective optimization attempts when the coverage of the agent is severely insufficient, thereby saving system resources and preventing new heating risks caused by insufficient optimization coverage.

[0076] Additionally, it should be noted that when the number of heating stations set up as described in S33 is not less than the preset threshold for the number of heating stations, the following content is included: Case 1: Determine the matching ratio based on the ratio of the set heating station to the heating station in the heating network. If the matching ratio is less than the preset matching ratio threshold, then determine that none of the weather types belong to the response optimization weather type. The set matching ratio refers to the ratio of the number of set heating stations to the total number of heating stations in the heating network; the preset matching ratio threshold refers to the pre-set ratio limit used to determine whether the overall participation of set heating stations in the entire network of heating stations meets the basic requirements of coordinated regulation.

[0077] If the heating network has multiple heating stations and the number of heating stations is relatively small, and the calculated matching ratio is lower than the preset matching ratio threshold, it indicates that the coverage density of the agent in the heating stations of the entire network is insufficient, and it is determined that none of the weather types belong to the response optimization weather types.

[0078] This situation is intercepted when the absolute number of heating stations meets the requirements but the overall coverage ratio is insufficient. The significance is that even if the number of heating stations reaches the minimum threshold, if their proportion in the whole network is too low, the effect of intelligent agent collaborative regulation on improving the reliability of the whole network heating is still very limited. Therefore, using the matching ratio as a second threshold can further ensure that the determination of the weather type for response optimization has practical application value.

[0079] Also includes the following: Case 2: If the set matching ratio is not less than the preset matching ratio threshold, the overlapping setting ratio is determined by the ratio of the set heat stations in the associated heat stations. If the overlapping setting ratio in the weather type is greater than the preset setting ratio threshold, the weather type with a daily average adjustment number less than the first adjustment number threshold is taken as the response optimization weather type. The overlapping setting ratio refers to the proportion of the number of heat stations that are associated with a certain weather type to the total number of heat stations associated with that weather type; the first adjustment frequency threshold refers to a pre-set daily average adjustment frequency limit used to distinguish between weather types with high and low adjustment demand intensity; the weather type with a daily average adjustment frequency less than the first adjustment frequency threshold refers to a weather type with relatively moderate adjustment demand and low risk of agent-coordinated adjustment.

[0080] If the proportion of heating stations set up in the associated heating stations under a certain weather type is high, exceeding the preset setting proportion threshold, and the average daily adjustment number of this weather type is lower than the first adjustment number threshold, then this weather type will be included in the response optimization weather type; if the average daily adjustment number of this weather type is high, exceeding the first adjustment number threshold, then it will not be included in the response optimization weather type.

[0081] In this case, when the overlap is sufficient, fine screening is performed by the average number of daily adjustments. The significance is that a high overlap ratio indicates that the agent has sufficient ability to cover the main heating stations for this weather type. However, a weather type with an excessively high adjustment frequency also means that the heating system is under greater pressure. If it is rashly included in the response optimization at this time, it may introduce new uncertainties and risks. Therefore, the adjustment frequency threshold is used as the final screening condition to ensure that only weather types with adjustment pressure within a controllable range are included in the response optimization weather types.

[0082] Case 3: If the overlap ratio of the weather types is not greater than the preset setting ratio threshold, determine whether the number of the response-optimized weather types is greater than the preset response-optimized weather type threshold. If yes, determine that none of the weather types belong to the response-optimized weather types. If no, then take the weather types among the above weather types that have a daily average number of adjustments less than the preset adjustment number and a number of associated heating stations less than the preset threshold for the number of associated heating stations as the response-optimized weather types. The preset adjustment number refers to the upper limit of the average daily adjustment number used to filter the response optimization weather type in case 3, and the preset adjustment number is less than the first adjustment number threshold; the preset associated heat station number threshold refers to the pre-set limit of the number of heat stations used to determine whether the scale of associated heat stations for a certain weather type is suitable for inclusion in response optimization; the preset response optimization weather type threshold refers to the pre-set limit value used to control the upper limit of the total number of response optimization weather types.

[0083] If the overlap ratio under a certain weather type does not exceed the preset setting ratio threshold, then first determine whether the number of currently identified response optimization weather types exceeds the preset response optimization weather type threshold; if it does not exceed the threshold, then further filter whether the average daily adjustment number of the weather type is lower than the preset adjustment number and whether the number of associated heat stations is lower than the preset associated heat station number threshold. If both conditions are met, then it is included in the response optimization weather type.

[0084] In cases where the overlap is insufficient, a more stringent dual-condition screening is employed. The significance of this is that a low overlap ratio indicates that the agent's coverage of the main regulating heat stations for that weather type is insufficient, and the collaborative regulation effect is somewhat limited. In this case, a lower regulation frequency threshold and a smaller requirement for the scale of associated heat stations are introduced to limit the application scenario of response optimization to weather types with relatively simple regulation needs. This maximizes the effective value of agent collaborative regulation under limited agent coverage conditions. At the same time, the upper limit on the total number of weather types for response optimization prevents the accumulation of collaborative risks caused by over-expansion.

[0085] Following the specific embodiment of S2, the heating stations are set as H1, H5, and H7, and the number of heating stations is set to 3.

[0086] S31 execution process: Confirm that the set of heating stations is {H1, H5, H7}, and the number of heating stations is 3.

[0087] S32 execution process: The preset threshold for the number of adjustments is set to 1.00 times / day. Statistics are compiled for associated heating stations under each weather type (average daily load adjustment times greater than 1.00 times / day): For W1, the associated heating stations are {H1, H3, H5, H7, H9}, a total of 5; for W2, the associated heating stations are {H2, H5, H7, H11}, a total of 4; for W3, the associated heating stations are {H4, H8, H14}, a total of 3; for W4, the associated heating stations are {H6, H10}, a total of 2; for W5, the associated heating station is {H13}, a total of 1; and for W6, the associated heating stations are {}, a total of 0.

[0088] The overlap ratio for each weather type is calculated as follows: Let P be the number of heat stations associated with a certain weather type, and Q be the total number of heat stations associated with that weather type. Then: Overlapping ratio = P ÷ Q W1: Overlap setting ratio = 3 ÷ 5 = 0.60; W2: Overlap setting ratio = 2 ÷ 4 = 0.50; W3: Overlap setting ratio = 0 ÷ 3 = 0.00; W4 and W5: The overlap ratio is set to 0.00 for both.

[0089] S33 execution process: The number of heating stations is set to 3, and the preset threshold for the number of heating stations is set to 2. If 3 is not less than 2, the judgment of "all do not belong to the response optimization weather type" will not be triggered, and the process will continue.

[0090] Set matching ratio = Number of heating stations ÷ Total number of heating stations in the network = 3 ÷ 20 = 0.15.

[0091] The preset matching ratio threshold is set to 0.10. 0.15 is not less than 0.10, which meets condition 2, so continue execution.

[0092] Case 2 judgment: The preset setting ratio threshold is set to 0.50.

[0093] W1: The overlap setting ratio of 0.60 is greater than 0.50, which meets the conditions of condition 2. The average number of adjustments of W1 per day (set to 1.80 times / day) is compared with the first adjustment threshold (set to 2.00 times / day): 1.80 < 2.00, which meets the conditions. W1 is included in the response optimization weather type.

[0094] W2: The overlap ratio is 0.50, which is not greater than 0.50 (equal to not meeting the "greater than" condition), proceed to condition 3. The number of currently identified response-optimized weather types is 1 (W1), and the preset response-optimized weather type threshold is set to 3. Since 1 is not greater than 3, continue the judgment. Compare the average daily adjustment frequency of W2 (set to 1.20 times / day) with the preset adjustment frequency value (set to 1.50 times / day): 1.20 < 1.50, satisfying the first condition; the number of associated heating stations of W2 is 4, and the preset threshold for the number of associated heating stations is set to 5. Since 4 < 5, satisfying the second condition, W2 is included in the response-optimized weather type.

[0095] W3: The overlap ratio is 0.00, which is no greater than 0.50. Proceed to condition 3. The current number of weather types for response optimization is 2 (W1, W2), which is no greater than the threshold 3. Continue the evaluation. Compare the average daily adjustment frequency of W3 (set to 0.60 times / day) with the preset adjustment frequency value of 1.50 times / day: 0.60 < 1.50, satisfying the first condition; the number of associated heating stations for W3 is 3, which is 3 < 5, satisfying the second condition. W3 is included in the response optimization weather type.

[0096] Ultimately, the optimized weather types for the response are W1, W2, W3, and W4.

[0097] Furthermore, the identification and matching data of the intelligent agent is determined based on whether the heat supply can meet the heating demand after the load is adjusted using the intelligent agent.

[0098] Furthermore, the method for determining the update strategy for the response optimization weather type in the aforementioned weather type is as follows: In this embodiment, based on the adjustment deviation of the thermal station set up by the intelligent agent in the response to the optimized weather type, it is determined how to update the response to the optimized weather type after the thermal station with the intelligent agent is updated. This allows for faster identification of the adjustment reliability of different intelligent agents while minimizing the risks of collaborative adjustment processing, laying the foundation for model updates and further promotion of the intelligent agents.

[0099] S41 uses the identification and matching data of the agent's load regulation demand in different response-optimized weather types to determine the time period in which the agent's set heating station cannot meet the heating demand in the response-optimized weather type, and takes it as the regulation deviation period. The identification and matching data of the intelligent agent refers to the matching result data determined based on whether the heat supply can meet the heating demand after load adjustment by the intelligent agent; the adjustment deviation period refers to the specific set of time periods under a certain response-optimized weather type where the heat supply of a certain heating station still fails to meet the heating demand after load adjustment by the intelligent agent.

[0100] Suppose that after the heating station H1 is adjusted by the intelligent agent in response to the optimized weather type W1, there are still some periods when the heating supply cannot meet the heating demand. These periods are marked as the adjustment deviation periods of H1 under W1, and the start and end times and duration of each deviation period are recorded.

[0101] This step provides raw data for calculating the proportion of adjustment deviation time in each response-optimized weather type by accurately identifying the adjustment deviation period of each set-up thermal station. Its significance lies in the fact that the identification of the adjustment deviation period directly reflects the actual effect of the agent's collaborative adjustment under different weather conditions. It is a key basis for evaluating the reliability of the agent's adjustment, thus providing a quantifiable evaluation basis for determining whether the response-optimized weather type needs to be updated and how to update it.

[0102] S42 determines the duration percentage of the adjustment deviation period of the set heating station in different response-optimized weather types based on the adjustment deviation period data of the set heating station in different response-optimized weather types, and takes the maximum value of the duration percentage of the adjustment deviation period of the set heating station in different response-optimized weather types as the adjustment deviation duration percentage of the set heating station. The duration of the adjustment deviation period refers to the ratio of the total duration of the adjustment deviation period of a certain thermal power station under a certain response optimization weather type to the total operating time of the agent under that weather type; the duration of the adjustment deviation period refers to the maximum value of the duration of the adjustment deviation period of a certain thermal power station among all response optimization weather types, reflecting the degree of adjustment deviation of the agent of the thermal power station under the most unfavorable weather conditions.

[0103] Suppose that the ratio of the total duration of the regulation deviation period of heating station H1 under W1 to the total operating time of the W1 agent is set to a certain value, another value under W2, and another value under W3. Then the maximum value among the three is taken as the proportion of the regulation deviation period of H1.

[0104] This step takes the maximum percentage of adjustment deviation duration in each response optimization weather type as the final adjustment deviation duration percentage for setting the heat station. Its significance lies in using the most unfavorable scenario as the evaluation benchmark, which can more comprehensively reflect the potential risks of agent collaborative adjustment, avoid underestimating the adjustment deviation risk under certain weather conditions by only considering the average level, and thus ensure that the determination of subsequent update strategies is based on the most conservative and safe evaluation results.

[0105] S43 determines the update strategy for the response optimization weather type based on the update matching results of the agent and the proportion of adjustment deviation time of different setting heat stations; The update matching result of the agent refers to the evaluation result of the degree of matching between the new version of the agent and the load adjustment demand under each response-optimized weather type after the agent completes the model update; the update strategy of the response-optimized weather type refers to the strategy scheme for dynamically adjusting and expanding the range of the response-optimized weather type after the agent completes the update.

[0106] Assuming that the proportion of adjustment deviation duration for each set of heating stations is low and the agent update matching results are good, the specific update strategy is determined based on the distribution of response deviation time periods. If the proportion of adjustment deviation duration for a certain set of heating stations exceeds the preset deviation proportion threshold, it indicates that the current agent's adjustment effect at that heating station is significantly insufficient, and it is determined that there is no need to perform response optimization weather type update processing.

[0107] This step, by comprehensively considering the agent's update matching results and setting the proportion of the heat station's adjustment deviation duration, achieves differentiated determination of the response optimization weather type update strategy. Its significance lies in the fact that different adjustment deviation distribution characteristics and update matching situations correspond to different update strategies. The refined strategy determination can avoid the scheduling risks brought about by unnecessary response optimization weather type updates, while ensuring that more weather types are included in the response optimization scope in a timely manner when conditions are ripe, thereby achieving continuous improvement and gradual promotion of the agent's collaborative adjustment capabilities.

[0108] It should be noted that if there are heating stations with an adjustment deviation ratio greater than the preset deviation ratio threshold, then the update strategy for the response optimization weather type in the weather type is determined to be that no update processing of the response optimization weather type is required. The preset deviation percentage threshold refers to a pre-set limit for judging whether the proportion of adjustment deviation duration of a certain heat station has reached an unacceptable level; no need to update the response optimization weather type means that when there are obvious defects in the current agent's collaborative adjustment effect, the existing response optimization weather type range remains unchanged and is not expanded for the time being, and is re-evaluated after the agent's adjustment reliability is improved.

[0109] Suppose that the proportion of adjustment deviation time of a certain heating station exceeds the preset deviation proportion threshold, it indicates that the adjustment deviation of the agent of that heating station is relatively high under certain weather conditions. In this case, continuing to expand the response optimization weather type may aggravate the heating risk. Therefore, the update strategy is determined to be that there is no need to perform response optimization weather type update processing.

[0110] This approach, by quickly intercepting the most unfavorable adjustment deviation, protects the response optimization weather type range from blindly expanding when the agent's adjustment effect is substandard. Its significance lies in using the extreme value of the proportion of adjustment deviation duration as the highest priority judgment condition, which can identify situations with significant risks in the agent's collaborative adjustment in the shortest time and make conservative strategy choices, thereby ensuring the adjustment reliability of the heating system during the agent's update transition period.

[0111] Additionally, it should be noted that if there are no heating stations in S43 where the adjustment deviation percentage is greater than the preset deviation percentage threshold, the following applies: S431 Determine whether there is a response-optimized weather type. If yes, proceed to step S432. If no, determine that the update strategy for the response-optimized weather type in the weather type is that no update processing of the response-optimized weather type is required, and use the above-mentioned method for determining the response-optimized weather type in the weather type to determine the response-optimized weather type. The existence of a response-optimized weather type refers to whether there is currently a weather type included in the scope of response-optimized weather types. If there is no response-optimized weather type, it means that the current system has not yet established a response optimization mechanism. In this case, the response-optimized weather type needs to be re-determined through the response-optimized weather type determination method in S3 to provide a basis for the execution of subsequent update strategies.

[0112] If there are no heat stations with a regulation deviation ratio exceeding the preset deviation ratio threshold, and there is already a weather type that can be optimized, then proceed to the next step of determining the regulation deviation coefficient.

[0113] This judgment ensures that the execution of subsequent update strategies has a valid target by confirming the existence of the response-optimized weather type. Its significance lies in the fact that if the response-optimized weather type does not currently exist, a foundation must first be established through the determination method to avoid logical errors caused by executing update strategies without a response-optimized weather type, thereby ensuring the integrity and executability of the overall process.

[0114] S432 is based on the proportion of adjustment deviation duration of the set heating stations under different weather types to determine the set heating stations with adjustment deviation periods in different response optimization weather types. Based on the number of set heating stations with adjustment deviation periods in different response optimization weather types, the adjustment deviation coefficient of the response optimization weather type is determined. It is then determined whether there is an optimization weather type with an adjustment deviation coefficient greater than a preset deviation coefficient threshold. If so, the update strategy for the response optimization weather type in the weather type is determined to be that no update processing of the response optimization weather type is required. If not, proceed to step S433. The term "heat stations with regulation deviation periods in different response-optimized weather types" refers to the set of heat stations whose regulation deviation duration is greater than zero (i.e., regulation deviation periods exist) under a certain response-optimized weather type; the term "regulation deviation coefficient for response-optimized weather type" refers to the ratio of the number of heat stations with regulation deviation periods in a certain response-optimized weather type to the total number of heat stations; the term "preset deviation coefficient threshold" refers to a pre-set proportional limit used to determine whether the overall regulation deviation risk under a certain response-optimized weather type is acceptable.

[0115] If there are many heat stations with adjustment deviations during a certain weather type under a certain response optimization, and the calculated adjustment deviation coefficient exceeds the preset deviation coefficient threshold, it indicates that the overall reliability of the agent's collaborative adjustment under this weather type is insufficient, and the update strategy is determined to be that no update processing of the response optimization weather type is required; if the adjustment deviation coefficients of all response optimization weather types do not exceed the preset deviation coefficient threshold, then proceed to the next step.

[0116] This judgment assesses the overall regulation deviation risk for each type of response-optimized weather, identifying response-optimized weather types with systemic regulation deficiencies. Its significance lies in the fact that even if the regulation deviation duration of a single heat station does not exceed the extreme threshold, if multiple heat stations under a certain weather type have periods of regulation deviation, the overall regulation reliability risk remains high. In this case, a conservative strategy should be adopted to suspend updates, thereby preventing the inappropriate expansion of the scope of response-optimized weather types due to the underestimation of systemic regulation deviation.

[0117] S433 determines whether the proportion of the set heating station with response deviation in the period is less than a preset proportion threshold. If so, the update strategy for the response optimization weather type in the weather type is determined to be that as long as there is a newly updated set heating station among the associated heating stations in the weather type and the average daily adjustment number in the weather type within the most recent preset duration is less than the first adjustment number threshold, it will be regarded as the response optimization weather type. If not, the update strategy for the response optimization weather type in the weather type is determined to be that as long as the newly updated set heating station among the associated heating stations in the weather type is greater than the preset set proportion threshold after the update and the average daily adjustment number in the weather type within the most recent preset duration is less than the first adjustment number threshold, it will be regarded as the response optimization weather type. The set of heat stations with response deviation periods refers to the set of heat stations with adjustment deviation periods in all response-optimized weather types; the preset proportion threshold refers to a pre-set boundary value used to determine whether the proportion of heat stations with adjustment deviations in all heat stations is at a low level; the newly updated heat stations refer to heat stations that have completed the update process in the most recent agent model update; the most recent preset duration refers to the time window used to evaluate the recent adjustment frequency of weather types; the updated proportion exceeding the preset setting threshold means that in the associated heat stations of a certain weather type, the proportion of the number of newly updated heat stations to the total number of associated heat stations exceeds the preset setting proportion threshold.

[0118] If the proportion of heat stations with response deviation periods is low, below a preset threshold, a lenient strategy is adopted: as long as there are newly updated heat stations among the associated heat stations for a certain weather type and the recent average daily adjustment frequency is lower than the first adjustment frequency threshold, it will be included in the response optimization weather type. If the proportion of heat stations with response deviation periods is high, not lower than the preset threshold, a strict strategy is adopted: it will only be included in the response optimization weather type if the proportion of newly updated heat stations among the associated heat stations exceeds the preset threshold and the recent average daily adjustment frequency is lower than the first adjustment frequency threshold.

[0119] This judgment distinguishes the applicable scenarios of the two update strategies by evaluating the proportion of heat stations with adjustment deviations. The significance is that when the proportion of heat stations with adjustment deviations is low, it indicates that the overall adjustment effect of the agent is good, and more lenient standards can be used to include more weather types in response optimization to accelerate the promotion progress. When the proportion is high, it indicates that there are still some problems with the agent's adjustment, and more stringent coverage ratio requirements are needed to control the risk of response optimization expansion, so as to achieve a dynamic balance between promotion speed and adjustment safety.

[0120] This embodiment achieves the complete determination of the response optimization weather type update strategy through S41 to S43. Its core value is reflected in three aspects: First, it conducts a comprehensive safety assessment of the agent's collaborative adjustment effect through the dual thresholds of adjustment deviation duration ratio and adjustment deviation coefficient; second, it designs a lenient strategy and a strict strategy for two scenarios: one with good overall adjustment effect and the other with certain deviation, respectively, to achieve differentiated adaptation of the update strategy; third, it organically combines the newly updated setting of the heat station coverage ratio with the recent adjustment frequency of the weather type as a dual condition for including the response optimization weather type, ensuring the scientific nature and safety of the update strategy execution.

[0121] Following the specific implementation of S3, the heat stations are set as {H1, H5, H7}, and the response optimization weather types are {W1, W2, W3, W4}.

[0122] S41 execution process: Based on the intelligent agent identification and matching data, the time periods of adjustment deviation for each set of thermal power stations in each response optimization weather type are statistically analyzed: The total duration of the adjustment deviation period for H1 under W1 is 12 hours, and the total operating time of the W1 agent is 200 hours; the total duration of the adjustment deviation period for H1 under W2 is 8 hours, and the total operating time of the W2 agent is 160 hours; there is no adjustment deviation period for H1 under W3; there is no adjustment deviation period for H1 under W4.

[0123] The total duration of the adjustment deviation period for H5 under W1 is 6 hours, and the total operating time of the W1 agent is 200 hours; the total duration of the adjustment deviation period for H5 under W2 is 5 hours, and the total operating time of the W2 agent is 160 hours; H5 has no adjustment deviation period under W3 and W4.

[0124] H7 has no adjustment deviation during the period under W1; H7 has no adjustment deviation during the periods under W2, W3, and W4.

[0125] S42 execution process: Let T_deviation be the total duration of the adjustment deviation period at a certain thermal power station under a certain weather type with optimized response, and T_total be the total operating time of the agent under this weather type. Then: The percentage of adjustment deviation duration = T_deviation ÷ T_total. The percentage of adjustment deviation duration for each weather type in H1 is as follows: W1: 12 ÷ 200 = 0.060; W2: 8 ÷ 160 = 0.050; W3: 0; W4: 0. The percentage of adjustment deviation duration in H1 (maximum value) = 0.060.

[0126] The percentage of adjustment deviation duration for each weather type in H5 is as follows: W1: 6 ÷ 200 = 0.030; W2: 5 ÷ 160 = 0.031; W3: 0; W4: 0. The percentage of adjustment deviation duration for H5 (maximum value) is 0.031.

[0127] The percentage of adjustment deviation time periods for each weather type in H7 is 0. The percentage of adjustment deviation time periods for H7 (maximum value) = 0.000.

[0128] S43 execution process: The preset deviation percentage threshold is set to 0.10. The adjustment deviation duration percentage of H1 is 0.060 < 0.10, the adjustment deviation duration percentage of H5 is 0.031 < 0.10, and the adjustment deviation duration percentage of H7 is 0.000 < 0.10. There are no heating stations with adjustment deviation duration percentages exceeding the preset deviation percentage threshold, so the process continues.

[0129] The current response optimization weather types are {W1, W2, W3, W4}. Since there are response optimization weather types, continue execution.

[0130] Calculate the adjustment deviation coefficient for each response-optimized weather type: Let M be the number of heating stations set up during the period with adjustment bias in a certain weather type of response optimization, and N be the total number of heating stations set up. Then: Adjustment deviation coefficient = M ÷ N, W1: Both H1 and H5 have adjustment deviation periods, M=2, N=3, adjustment deviation coefficient = 2÷3 = 0.667; W2: Both H1 and H5 have periods of adjustment deviation, M=2, N=3, adjustment deviation coefficient = 2÷3 = 0.667; W3, W4: No heating station is set up and there are periods of adjustment deviation, M=0, N=3, adjustment deviation coefficient = 0.

[0131] The preset deviation coefficient threshold is set to 0.70. The adjustment deviation coefficient of W1 is 0.667 < 0.70, the adjustment deviation coefficient of W2 is 0.667 < 0.70, and the adjustment deviation coefficients of W3 and W4 are 0 < 0.70. The adjustment deviation coefficients of all weather types that respond to optimization do not exceed the preset deviation coefficient threshold, so execution continues.

[0132] The number of heating stations with response deviations during the specified periods is 2, which accounts for 2% of the total number of heating stations (3). The percentage of these 2% is 2 ÷ 3 = 0.667.

[0133] The preset ratio threshold is set to 0.70. 0.667 < 0.70, which meets the condition of "less than the preset ratio threshold". Therefore, the update strategy for the response optimization weather type is determined to be a lenient strategy: as long as there is a newly updated set heat station among the associated heat stations of a certain weather type, and the average daily adjustment number of that weather type is less than the first adjustment number threshold (2 times / day) within the most recent preset duration (set to 30 days), it will be regarded as the response optimization weather type.

[0134] Example 2 On the other hand, this application provides a computer system that stores a computer program. When the computer program is executed in the computer, it causes the computer to execute the above-mentioned intelligent agent-based heating network scheduling method.

[0135] Example 3 On the other hand, this application provides a computer program product, characterized in that the computer program product stores instructions, which, when executed by a computer, cause the computer to implement the above-described agent-based heating network scheduling method.

[0136] In this embodiment of the invention, the term "multiple" refers to two or more, unless otherwise explicitly defined. The terms "install," "connect," and "fix" should be interpreted broadly. For example, "connect" can mean a fixed connection, a detachable connection, or an integral connection. Those skilled in the art can understand the specific meaning of the above terms in this embodiment of the invention based on the specific circumstances.

[0137] In the description of the embodiments of the present invention, it should be understood that the terms "upper" and "lower" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the embodiments of the present invention and simplifying the description, and do not indicate or imply that the device or unit referred to must have a specific orientation or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the embodiments of the present invention.

[0138] In the description of this specification, the terms "an embodiment," "a preferred embodiment," etc., refer to a specific feature, structure, material, or characteristic described in connection with that embodiment or example, which is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0139] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. For those skilled in the art, the embodiments of the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of the present invention should be included within the protection scope of the embodiments of the present invention.

Claims

1. A heating network scheduling method based on intelligent agents, characterized in that, Specifically, it includes: Using load regulation data, identify load regulation anomaly data under different weather types. Based on the load regulation anomaly data, identify the load regulation anomaly type and associated abnormal heat stations under different weather types. Utilize the degree of overlap of associated abnormal heat stations under different load regulation anomaly types to determine the update setting strategy for the heat station's intelligent agent. The setting data of the agent is determined by using the agent's update setting strategy, and the overlap between the agent's setting data and the load adjustment of the heat station under different weather types is combined to determine the response optimization weather type in the weather type. By utilizing the identification and matching data of agents regarding load adjustment needs in different response-optimized weather types, and combining this with the agent's update matching results, the update strategy for the response-optimized weather type in the aforementioned weather type is determined.

2. The heating network scheduling method based on intelligent agents as described in claim 1, characterized in that, The abnormal load regulation data is determined based on whether the heat supply can meet the heating demand after load scheduling processing under the specified weather type.

3. The heating network scheduling method based on intelligent agents as described in claim 1, characterized in that, The method for determining the type of load regulation anomaly under the aforementioned weather type is as follows: Based on the load regulation anomaly data under the aforementioned weather type, determine the number of times that the heating station cannot meet the heating demand after load scheduling under the aforementioned weather type, and use this as the number of regulation deviations. The number of adjustment deviations is used to determine the associated abnormal heat stations under the weather type; Based on the associated abnormal heat station data under the aforementioned weather type, the type of load regulation anomaly under the aforementioned weather type is determined.

4. The agent-based heating network scheduling method as described in claim 3, characterized in that, The associated abnormal heat station is the heat station whose daily average number of adjustment deviations under the weather type is greater than a preset deviation number threshold.

5. The heating network scheduling method based on intelligent agents as described in claim 4, characterized in that, Based on the associated abnormal heat station data under the aforementioned weather type, the type of load regulation anomaly under the aforementioned weather type is determined, specifically including: Based on the number of associated abnormal heat stations under the aforementioned weather type, a preset abnormality type corresponding to the number of associated abnormal heat stations is determined; The load adjustment anomaly type under the weather type is determined by using the corresponding preset anomaly type.

6. The heating network scheduling method based on intelligent agents as described in claim 5, characterized in that, The load regulation anomaly types include three types: Type I, Type II, and Type III. The degree of anomaly in load regulation of Type I anomaly type is greater than that of Type II anomaly type, and the degree of anomaly in load regulation of Type II anomaly type is greater than that of Type III anomaly type.

7. The heating network scheduling method based on intelligent agents as described in claim 1, characterized in that, The method for determining the update setting strategy of the intelligent agent of the heating station is as follows: Based on the load adjustment anomaly type of the aforementioned weather type, determine the number of weather types under different load adjustment anomaly types; Based on the degree of overlap of associated abnormal heat stations under different load regulation anomaly weather types, the weather type to which the heat station belongs to the associated abnormal heat station is determined, and it is taken as the associated abnormal weather type of the heat station. By utilizing the number of weather types under different load regulation anomaly types and the associated abnormal weather types of different heating stations, the update setting strategy of the intelligent agent of the heating station is determined.

8. The heating network scheduling method based on intelligent agents as described in claim 7, characterized in that, If the number of weather types under a certain abnormality type is greater than the preset threshold for the number of weather types, then if the heating station has an adjustment deviation in weather types above the second preset number, then the setting process of the intelligent agent will be performed.

9. The agent-based heating network scheduling method as described in claim 1, characterized in that, The method for determining the update strategy for the response optimization weather type in the aforementioned weather type is as follows: By using the identification and matching data of agents' load regulation needs in different response-optimized weather types, the time periods in which the heating station set by the agent cannot meet the heating demand in the response-optimized weather type are identified and these periods are taken as regulation deviation periods. Based on the adjustment deviation time period data of the set heating station in different response-optimized weather types, the duration ratio of the adjustment deviation time period of the set heating station in different response-optimized weather types is determined, and the maximum value of the duration ratio of the adjustment deviation time period of the set heating station in different response-optimized weather types is taken as the adjustment deviation duration ratio of the set heating station. Based on the update matching results of the agent and the proportion of adjustment deviation duration of different setting heat stations, the update strategy for response optimization weather type in the weather type is determined.

10. A computer system having a computer program stored thereon, wherein when the computer program is executed in the computer, it is characterized in that, The computer is instructed to execute the agent-based heating network scheduling method according to any one of claims 1-9.

Images