An agent-based long-distance pipeline network operation management method and system

Abstract

This invention provides an operation management method and system for long-distance pipeline networks based on intelligent agents, belonging to the field of intelligent agent technology. Specifically, it includes: designating pump stations with intelligent agents as intelligent agent pump stations; determining the coordinated adjustment strategy for intelligent agent pump stations based on overlapping data between intelligent agent pump stations and operating pump stations, excluding power outage data of pump stations other than intelligent agent pump stations; performing coordinated adjustment processing of intelligent agent pump stations based on the coordinated adjustment strategy to obtain operational results; and determining the control and management method for setting intelligent agent objectives based on operational results under different combinations of coordinated adjustment of intelligent agent pump stations, thereby improving the operational reliability of long-distance pipeline networks.

Description

Technical Field

[0001] This invention belongs to the field of intelligent agent technology, and in particular relates to an operation and management method and system for long-distance pipeline networks based on intelligent agents. Background Technology

[0002] The reliable operation of long-distance pipelines depends on pumping stations. However, these pumping stations are widely distributed and often employ differentiated power supply methods. A power outage at any pumping station can affect the transmission reliability of the entire long-distance pipeline network. Therefore, how to construct intelligent agents in pumping stations and utilize these agents for coordinated scheduling of pumping stations during power outages to improve the operational reliability of long-distance pipelines has become an urgent technical problem to be solved.

[0003] Specifically, this application provides a method and system for the operation and management of long-distance pipeline networks based on intelligent agents. Summary of the Invention

[0004] To achieve the objectives of this invention, the following technical solution is adopted: Specifically, this application provides an operation and management method for long-distance pipeline networks based on intelligent agents, which includes: S1 uses the operation data of the long-distance pipeline network to determine the operating pump stations in the long-distance pipeline network at different time periods. Based on the operation data of the pump stations at different time periods, it determines the setting strategy of the intelligent agents of the pump stations based on power outage data. It uses the setting strategy to perform the setting process of the intelligent agents. Based on the distribution data of the intelligent agents and combined with the matching degree of the pump stations with intelligent agents in the power outage event, it determines when the control processing of the setting target of the intelligent agents needs to be carried out, and then proceeds to the next step. S2 takes the pump station with intelligent agents as the intelligent agent pump station, and determines the collaborative adjustment strategy of the intelligent agent pump station based on the overlapping data of the intelligent agent pump station and the operating pump station, excluding the power outage data of the pump station other than the intelligent agent pump station, and performs collaborative adjustment processing of the intelligent agent pump station based on the collaborative adjustment strategy to obtain the operation result. S3 determines the control and management methods for the setting objectives of the intelligent agent based on the operating results under the coordinated adjustment combination of different intelligent agent pump stations.

[0005] The beneficial effects of this invention are as follows: Based on the operational pump station data at different times, the system determines the configuration strategy for intelligent agents based on power outage data. According to the "redundancy stress" of the overall pipeline network operation, it dynamically decides which pump stations to prioritize for intelligent agents. In the event of an unexpected power outage, the intelligent agents coordinate the station's energy storage (such as UPS, emergency power supply) to maintain the pump station's operation for a short period or shut it down in an orderly manner. This provides the dispatch center with valuable "golden buffer time" to remotely or automatically switch to other backup pump stations, thereby preventing heating interruptions or drastic fluctuations in pipeline network pressure due to a sudden power outage of a single pump station, and ensuring the overall stability and resilience of the heating system.

[0006] Based on the operational results of different intelligent agent pump station collaborative adjustment combinations, the management and control methods for intelligent agent setting targets are determined. This involves screening out "reliable pump stations," namely those intelligent agents that have been fully validated under various challenging scenarios while in shutdown mode. The system also assesses the risks of non-intelligent pump stations by considering the actual impact of power outages during these "stress tests" (especially the extent to which they extend system adjustment time). Finally, by combining the "reliability saturation of deployed intelligent agents" with the "risk exposure of undeployed pump stations," a decision is made on whether to maintain the existing deployment strategy and continue validation, or to list certain types of pump stations as new deployment targets. This approach achieves reliable identification of intelligent agent setting targets while minimizing the impact on long-distance pipeline networks.

[0007] Furthermore, the operating pump station in the time period refers to the pump station that is in operation during the time period.

[0008] Furthermore, the method for determining the setting strategy of the intelligent agent based on power outage data for the pumping station is as follows: The percentage of pump stations in operation during different time periods is determined by using data on operating pump stations during those different time periods. Based on the stated proportions, determine the time periods within the stated time period that require reliable operation. Based on the distribution data of the reliable operation demand period, the setting strategy of the intelligent agent of the pumping station based on power outage data is determined.

[0009] Furthermore, the method for determining the control and management method of the setting target of the intelligent agent is as follows: Based on the operational results of the coordinated regulation combination of intelligent pump stations, determine the coordinated operation period under the coordinated regulation combination; Based on the cooperative operating segments under different combinations of cooperative adjustment, the identified reliable pump stations in the intelligent pump station are determined. Based on the identified reliable pump station data and the power outage impact data of pump stations other than the intelligent agent pump station during the collaborative operation period, a management and control method for the setting target of the intelligent agent is determined.

[0010] In a second aspect, the present invention provides a computer system comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-described method for the operation and management of a long-distance pipeline network based on an intelligent agent when running the computer program.

[0011] Other features and advantages will be set forth in the following description, and the objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0012] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0013] 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.

[0014] Figure 1 This is a flowchart of an agent-based operation and management method for long-distance pipeline networks; Figure 2 This is a flowchart illustrating the method for determining the setting strategy of the intelligent agent in a pumping station based on power outage data. Figure 3 It is a flowchart for determining the control and management processes required to set up intelligent agents. Detailed Implementation

[0015] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0016] Example 1 like Figure 1 As shown, this application provides an operation and management method for long-distance pipeline networks based on intelligent agents, specifically including: S1 uses the operation data of the long-distance pipeline network to determine the operating pump stations in the long-distance pipeline network at different time periods. Based on the operation data of the pump stations at different time periods, it determines the setting strategy of the intelligent agents of the pump stations based on power outage data. It uses the setting strategy to perform the setting process of the intelligent agents. Based on the distribution data of the intelligent agents and combined with the matching degree of the pump stations with intelligent agents in the power outage event, it determines when the control processing of the setting target of the intelligent agents needs to be carried out, and then proceeds to the next step. S2 takes the pump station with intelligent agents as the intelligent agent pump station, and determines the collaborative adjustment strategy of the intelligent agent pump station based on the overlapping data of the intelligent agent pump station and the power outage data of the operating pump station excluding the intelligent agent pump station. Based on the collaborative adjustment strategy, the intelligent agent pump station is processed for collaborative adjustment to obtain the operation result. S3 determines the operational results under different coordinated adjustment combinations of intelligent agent pump stations, and combines the joint operational data of the coordinated adjustment combination with different operating pump stations to determine the control and processing methods for the intelligent agent's setting objectives.

[0017] Furthermore, the operating pump station in the time period refers to the pump station that is in operation during the time period.

[0018] Furthermore, such as Figure 2 As shown, the method for determining the setting strategy of the intelligent agent based on power outage data in the pumping station is as follows: For long-distance heating networks under critical tasks such as winter heating, a preventative deployment strategy of "intelligent agents" linked to system operation status is designed. Here, "intelligent agent" refers to an emergency control unit integrated within the pumping station, possessing local computing, energy storage management, and coordination with the central dispatch center. This strategy aims to dynamically determine which pumping stations should be prioritized for intelligent agent deployment based on the overall network's "redundancy stress." Its fundamental purpose is to utilize intelligent agents to coordinate local energy storage (such as UPS or emergency power supplies) to maintain pumping station operation briefly or orderly shutdown in the event of an unexpected power outage. This provides the dispatch center with valuable "golden buffer time" to remotely or automatically switch to other backup pumping stations, thereby preventing heating interruptions or drastic fluctuations in network pressure due to sudden power loss at a single pumping station, and ensuring the overall stability and resilience of the heating system.

[0019] This solution follows a decision chain of "perceiving system status -> assessing risk level -> accurately matching defense resources." The system analyzes historical operational data to identify "tight" periods where most pumping stations require operation and system redundancy is extremely low. If this "tight" state is the norm, it indicates that the pipeline network is in a long-term high-risk operating mode, highly dependent on the continuous operation of any pumping station. In this case, a more aggressive (stricter) defense strategy is needed, namely, deploying agents to more pumping stations that have experienced power outages to broadly enhance the network's buffer capacity. Conversely, if the "tight" state is an occasional occurrence, limited agent resources can be concentrated on a few "weak" pumping stations with the highest power outage risk and the most historical power outages. The core idea is that the breadth of agent deployment is directly proportional to the system's inherent vulnerability.

[0020] S11 uses data on operating pump stations in different time periods to determine the percentage of operating pump stations in different time periods; Operating pump stations: Pump stations that are actually in operation, such as pressurization or circulation, during the statistical period (e.g., every 15 minutes or 1 hour), rather than just being in a power-on standby state.

[0021] Percentage of operating pump stations: The percentage of pump stations actually in operation within a specific time period out of the total number of pump stations under the jurisdiction of the pipeline network. For example, in a heating ring network consisting of 12 pump stations, if 10 are in operation at a certain time period, the percentage is 83.3%.

[0022] This step is fundamental to assessing system load and redundancy levels. During peak heating seasons in severe cold periods, to meet the heat demands of distant users, the pipeline network often needs to activate the vast majority of pump stations to maintain sufficient flow and pressure differential, resulting in a high operational occupancy rate. A high occupancy rate directly means that the system has scarce available online backup pump station resources, reducing overall scheduling flexibility and placing extremely high demands on the stability of the current operating unit combination. The unexpected shutdown of any operating pump station can lead to pressure imbalances due to flow redistribution, affecting heating quality and even causing safety issues. Therefore, this occupancy rate is the primary indicator for measuring the "tightness" of the pipeline network's operation.

[0023] S12 determines the operational reliability demand period within the time period based on the stated quantity proportion; Preset percentage threshold: An empirical value set according to the redundancy of pipeline design and safe operation procedures, used to define the "high load, low redundancy" state, for example, 75%.

[0024] Reliable operating demand periods: The set of all periods that meet the condition that "the percentage of operating pump stations > the preset percentage threshold (e.g., 75%)". These periods are the "critical risk windows" where the pressure on heating supply is greatest and the system is most vulnerable.

[0025] This step enables the "precise identification and extraction of high-risk periods." It avoids general daily average analysis, focusing instead on the most demanding times for continuous and stable operation, typically the morning and evening heating peaks and periods of extreme cold. During these windows, the pipeline network is like a fully drawn bowstring; a break in any "string" (operating pump stations) could have severe consequences. Identifying these periods defines clear "key protection periods" for subsequent defense resource (agent) allocation. All decisions will serve to ensure the "uninterrupted power supply" or "orderly withdrawal after power outage" of operating pump stations during these critical periods.

[0026] S13 uses the distribution data of the reliable operation demand period to determine the setting strategy of the intelligent agent of the pumping station based on power outage data.

[0027] It is understood that the period of reliable operation requirement refers to the period in which the proportion of the quantity is greater than a preset proportion threshold.

[0028] Specifically, based on the distribution data of the reliable operating demand period, the setting strategy of the intelligent agent of the pumping station based on power outage data is determined, including: If the proportion of reliable operation demand periods in the time period is greater than the preset time period number proportion threshold, then the setting strategy of the intelligent agent of the pump station based on power outage data is determined to be to set the pump station with more than the preset number of power outages as the intelligent agent setting target. If the proportion of reliable operation demand periods in the time period is not greater than the preset time period quantity proportion threshold, then the setting strategy of the intelligent agent of the pump station based on power outage data is determined to be to set the pump station with more than the second preset number of power outages as the intelligent agent setting target.

[0029] It should be noted that the preset number of times threshold is less than the second preset number of times threshold.

[0030] Percentage of Reliable Demand Periods: Within a defined analysis period (such as a past heating season), the total duration of reliable demand periods accounts for the total duration of the analysis period. For example, if 950 hours out of a quarter of 2160 hours belong to reliable demand periods, the percentage is approximately 44%.

[0031] Preset time period quantity ratio threshold: A decision threshold for judging whether a "high-risk window" is normal or occasional, for example, 40%.

[0032] Preset duration threshold and second preset duration threshold: used to filter the power outage history threshold of pump stations that need to deploy intelligent agents, and the preset duration threshold < the second preset duration threshold (e.g.: preset = 2 minutes, second preset = 5 minutes).

[0033] This step serves as a decision-making bridge connecting the "system risk model" with "specific defensive actions," embodying the principle that "the higher the risk, the stricter the defense."

[0034] Scenario 1 (High-Risk Normalization, Strict Strategy): If the proportion of reliable demand periods is very high (>40%), it indicates that the pipeline network is in a state of "tight tension" for a long time, and systemic vulnerability is the main problem. In this case, even if a pumping station has a historical number of power outages greater than a preset threshold (e.g., >10 times), such short-term power outages are enough to trigger a chain reaction of problems when operating at a high proportion. Therefore, it is necessary to adopt strict deployment standards (low threshold) to include more pumping stations with a history of short-term power outages in the coverage of the intelligent agent, and to broadly enhance the "power outage resistance" of all nodes in the network.

[0035] Scenario 2 (High-Risk, Intermittent Issues, Precise Strategy): If this proportion is low (≤40%), it indicates that the pipeline network operates with ample redundancy most of the time. The main risk comes from a few "stubborn" pump stations with inherently unreliable power supply and a historical record of more than the second preset threshold for power outages (e.g., >20 times). In this case, a lenient deployment standard (high threshold) should be adopted, concentrating resources to prioritize equipping these "weakest links" with intelligent agents to address the most prominent single-point problems and maximize investment returns.

[0036] Final action: After determining the strategy, the system will retrieve the historical power outage records of all pump stations, filter out the list of pump stations whose number of power outages exceeds the corresponding threshold, and list them as "intelligent agent setting targets" for priority installation or upgrade.

[0037] Specific example of coherence: Suppose a long-distance heating pipeline network in a northern city includes 10 booster pump stations. Analyze the operating data from the past month (720 time periods, granularized by hour).

[0038] S11 calculation percentage: Count the number of pump stations operating each hour. For example, during the nighttime off-peak period, 4-5 pump stations usually operate (accounting for 40%-50%), and during the daytime peak period, 8-9 pump stations operate (accounting for 80%-90%).

[0039] S12 identifies key windows: a preset percentage threshold of 75% is set. Statistics show that approximately 8 hours per day (morning, noon, and evening peak hours) have ≥8 pump stations operating (percentage ≥80%), meeting the criteria. A total of 240 time periods per month are marked as "reliable operation demand periods".

[0040] S13 Decision-Making and Deployment Strategy: The percentage of time periods with reliable operational requirements is approximately 33% (240 / 720).

[0041] Set the preset time period quantity ratio threshold to 40%.

[0042] Judgment: 33% < 40%, which falls under the "high-risk occasional" scenario.

[0043] Decision: A lenient strategy will be adopted. The agent setting strategy is as follows: pump stations with a history of single power outages lasting more than 5 minutes will be selected as the target for agent setup.

[0044] Execution: The system query of the power outage database revealed that pump stations #3 and #7 had experienced more than 20 power outages in the past year, while other pump stations had experienced fewer than 20. Therefore, it was determined that pump stations #3 and #7 would be prioritized for deployment / upgrading of intelligent agents. When these pump stations experience another power outage, the intelligent agents will immediately take over, utilizing the station's energy storage to maintain circulation or execute a smooth shutdown procedure, while simultaneously alerting the dispatch center. This will allow the center at least 5 minutes of critical operation time to remotely start backup pump stations such as #4 and #8.

[0045] This embodiment demonstrates a "resilience investment" decision-making model that precisely links cost-effectiveness with system risk, achieving optimal allocation of defense resources: enhancing the overall resilience of the heating system: by adding a "power outage buffer layer" to key nodes, "hard faults" that could otherwise lead to heating interruptions are transformed into "soft faults" that can be scheduled and dealt with, significantly enhancing the ability of long-distance pipelines to cope with external shocks such as power disturbances. The strategy is entirely based on historical operation and power outage data, and can automatically reassess and adjust according to changes in pipeline load patterns, commissioning of new pumping stations, and external power grid upgrades, possessing continuous optimization capabilities.

[0046] Furthermore, the distribution data of the agents includes the number of pump stations where agents exist.

[0047] Specifically, such as Figure 3 As shown, the control processing required for setting the intelligent agent's objectives includes: In the phased deployment of intelligent agents in long-distance heating pipeline networks, a dynamic evaluation and braking mechanism should be established. Its core objective is to determine whether the current deployment scope of the intelligent agents is sufficiently effective, thus avoiding the trap of "over-deployment," i.e., continuing to invest costs under conditions of diminishing marginal returns. This approach does not halt deployment, but rather uses "management processing" to prudently assess whether the original, potentially aggressive deployment strategy should continue, or to optimize the strategy if the results are unsatisfactory.

[0048] This solution follows a path of "assessing coverage breadth -> verifying protection effectiveness -> comprehensive decision on whether to intervene in the original plan." The system examines the current situation from two dimensions: first, the static coverage ratio of the agents (how many pump stations are protected); and second, the dynamic protection effectiveness of the agents in historical real-world scenarios (power outage events) (whether unprotected pump stations frequently become points of failure). If the coverage is broad and the protection effect is good, "control" is triggered, suspending or re-evaluating the original deployment strategy to prevent resource waste; if the coverage is insufficient or the effect is poor, it proves that the original strategy still needs to be firmly implemented, and no control is required. The core insight is: when there are too many agent deployment targets, a "cooling-off period" is needed to assess and confirm that the investment is still focused on the real risk gaps.

[0049] S21 Based on the distribution data of the intelligent agents, determine the pump stations where intelligent agents exist, and designate the pump stations where intelligent agents exist as intelligent agent pump stations; The above steps include the following: S211 Determine whether the proportion of intelligent agent pump stations in the pump station is greater than a preset proportion threshold. If yes, determine that intelligent agent setting target control processing is required. If no, proceed to step S212. Intelligent pump station: A pump station that has completed the installation and commissioning of intelligent systems and has local emergency control capabilities.

[0050] The percentage of pump stations with deployed intelligent agents: This is the proportion of pump stations with deployed intelligent agents to the total number of pump stations. This is a core indicator for measuring the breadth of protection resource investment.

[0051] Preset percentage threshold: A critical value that indicates "the deployment scope may be broad enough", such as 70%.

[0052] This step is a "rapid initial screening for deployment saturation." If the proportion of intelligent agent pump stations is already very high (over 70%), it means that the vast majority of pump stations are already protected. From an investment perspective, continuing to deploy the remaining few pump stations according to the original strategy (possibly based on a short power outage duration threshold) may not yield marginal security benefits far below the cost. At this point, the system has sufficient reason to trigger "control processing," that is, to suspend the automatic execution of the original strategy and instead initiate a special assessment to decide whether to continue covering the last few points or to redirect resources to other areas such as optimizing the performance of existing intelligent agents. This prevents a decline in investment efficiency due to inertia in the later stages of deployment.

[0053] S212 takes the pump stations other than the intelligent agent pump station as other pump stations, and determines whether there are pump stations with a history of power outages among the other pump stations. If so, proceed to step S22; otherwise, determine that intelligent agent setting target management processing is required. Other pumping stations: Pumping stations where intelligent agents have not yet been deployed.

[0054] Pumping stations with a history of power outages: These are pumping stations listed under "Other Pumping Stations" that have a history of power outages.

[0055] This step identifies "clear remaining risk points." If the agent coverage is not too high (not exceeding the threshold), but all undeployed pump stations have never experienced a power outage, then these stations are historically considered "low-risk" sites. In this case, continuing to target them according to the original strategy (possibly based on historical power outage records) is no longer justifiable. Therefore, "control measures" should be triggered to reassess the necessity of deployment at these "clean" sites, potentially adjusting the strategy (such as extending the power outage duration threshold or postponing deployment). Conversely, if there are indeed pump stations with a "previous record" (having experienced power outages) among the uncovered sites, it proves that the risk still exists, requiring the next step to examine the actual effectiveness of the existing agent deployment.

[0056] S22 determines, based on the matching degree of the intelligent pump station in the power outage event, the pump stations that do not belong to the intelligent pump station in the power outage event, and regards them as the affected pump stations. It is understood that the above steps include the following: S221 Determine whether the pump station is affected in different power outage events. If so, determine that no control processing of the intelligent agent's setting target is required. If not, proceed to step S222. Power outage event: refers to any power outage accident of a pumping station that occurs in the pipeline network and is recorded by the system.

[0057] Impact on pumping stations: In a specific power outage event, the pumping station that experiences the power outage happens to be one without deployed agents (i.e., an "agent pumping station"). This signifies a "defense failure"—the agent network failed to cover the actual risk that occurred.

[0058] This step shifts from static coverage to dynamic verification, checking whether the agent deployment is truly "targeting the right areas." If, in every historical power outage event, the pumping stations affected by the outage are unprotected "affected pumping stations," this indicates a serious problem: the existing agent deployment is spatially mismatched with the actual power outage risk. The agents are not protecting the pumping stations that are truly prone to problems. This strongly suggests that the original deployment strategy (such as filtering based on the number of power outages) may be ineffective. The most urgent need is not "control" on whether to continue deployment, but to immediately correct the deployment direction. Therefore, "no control required" means that the original strategy must be continued or even strengthened to quickly plug this fatal protection loophole.

[0059] S222 defines power outage events that affect the pump station as power outage events, and determines whether the proportion of the number of power outage events in the power outage events is greater than a preset power outage event proportion threshold. If yes, it is determined that no control processing of the intelligent agent's setting target is required; otherwise, it proceeds to step S23. Events affecting power outages: An event that occurs when a power outage occurs and the pumping station is not protected by an agent.

[0060] Percentage of events affecting power outages: The proportion of "events affecting power outages" among all historical power outage events.

[0061] Preset power outage event percentage threshold: A threshold for judging whether protection failures are frequent, such as 60%.

[0062] This step assesses the frequency of protection failures. If most power outages (over 60%) occur at unprotected pumping stations, it indicates that the existing agent network's protection effectiveness is very low, leaving a large number of risks exposed. In this case, "control" (i.e., limiting deployment) cannot be implemented because the gap is huge, and deployment must continue to expand the protection scope. Only when the proportion of "impacting events" is low can it be said that the existing agent network has largely covered the main risk points, and the final comprehensive trade-off (S23) can be entered.

[0063] Based on the data from the intelligent pump station and the impact of the power outage event on the pump station, S23 determines whether it is necessary to perform control processing on the intelligent agent's set target.

[0064] Furthermore, in the above steps, the influence coefficient is determined based on the proportion of intelligent agent pump stations in the pump station and the proportion of the number of power outage events in the power outage events. When the influence coefficient is greater than the preset influence coefficient threshold, it is determined that no control processing of the intelligent agent setting target is required.

[0065] Furthermore, the influence coefficient is related to the proportion of intelligent agent pump stations in the pumping station and the proportion of the number of power outage events in the power outage events. The smaller the proportion of intelligent agent pump stations in the pumping station and the larger the proportion of the number of power outage events in the power outage events, the larger the influence coefficient.

[0066] Furthermore, when there is no need to manage the setting goals of the intelligent agent, the setting of the intelligent agent will still be carried out according to the original intelligent agent setting strategy based on power outage data.

[0067] Impact coefficient: A comprehensive quantitative indicator, typically calculated as follows: Impact coefficient = (1 - Proportion of intelligent pumping stations) + Percentage of events affecting power outages W1 and W2 are the weights. This formula ensures that the lower the coverage and the higher the protection failure rate, the larger the coefficient value. A preset impact coefficient threshold is set as the critical value for the final decision.

[0068] This step involves a "fine-grained trade-off between cost (coverage) and benefit (protection effectiveness)." It deals with an intermediate state: the agent's coverage is neither particularly broad nor the number of protection failures is particularly high. By calculating the "impact coefficient," the system quantifies the "severity of residual risk and protection inadequacy."

[0069] If the impact coefficient is very high (>threshold): This indicates that at least one of the problems of "low coverage" and "high failure rate" is prominent, and the system as a whole is still in a high-risk, low-protection state. Therefore, no control measures are needed, and the original deployment strategy should continue to be implemented to improve coverage and protection effectiveness as soon as possible.

[0070] If the impact coefficient is small (≤ threshold): it indicates that the coverage of intelligent agents has reached a certain level (the proportion is not particularly small), and the protection effect is acceptable (the proportion of failure events is not high). The current deployment has entered the "diminishing marginal returns" stage. At this time, it is determined that control measures are needed.

[0071] Specific examples Suppose a heating network has 10 pumping stations, and 5 smart entities have been deployed (smart entity pumping stations account for 50%). There have been 10 power outages in the past year.

[0072] S21 execution: S211: The proportion of intelligent pump stations is 50% < the preset threshold of 70%, proceed to S212.

[0073] S212: Of the 5 "other pumping stations" (not yet deployed), 3 have a history of power outages. Therefore, this is referred to S22.

[0074] S22 execution: Analysis of 10 power outage events revealed that 7 of them occurred at pump stations where no agents were deployed (i.e., 7 were "power outage events affecting power outages").

[0075] S221: Not every event is an impactful event (3 of them occurred at the protected pump station), proceed to S222.

[0076] S222: Percentage of power outage events = 7 / 10 = 70%. Assuming the preset threshold for power outage event percentage is 60%, since 70% > 60%, according to S222, the system should determine "no control is needed," and continue deployment. However, for demonstration purposes, we assume the threshold is 80%, and then proceed to S23.

[0077] S23 execution (assuming this branch is entered): Given: Intelligent pump stations account for 50% of the total, and events affecting power outages account for 70% of the total.

[0078] Let the formula be: Influence coefficient = (1 - 50%) + 70% = 120%. Let the preset threshold for the influence coefficient be 100%.

[0079] Judgment: 120% > 100%. Therefore, it was ultimately determined that "no control processing of the agent's setting target is required." The system assessment concluded that, with the current 50% coverage, as many as 70% of power outage events still occur in unprotected areas, indicating a significant remaining risk. Therefore, "control" will not be triggered, and the existing agent setting strategy based on power outage data will continue to be followed to select and deploy the next batch of agents until more high-risk pump stations are covered.

[0080] This embodiment demonstrates a crucial "control and management solution" in the process of intelligent transformation, preventing "deployment for the sake of deployment": it introduces the "control and management processing" decision node, forcing the system to review whether the investment has truly translated into security benefits after reaching a certain scale, avoiding blind expansion of projects. The decision is not only based on static coverage, but more importantly, on verifying the dynamic protection effect based on historical fault data. By providing a scientific basis for switching between "continuing to expand coverage" and "optimizing existing capabilities", it ensures that the total project budget can generate the greatest overall security benefits.

[0081] Furthermore, the method for determining the coordinated adjustment strategy of the intelligent pump station is as follows: For long-distance heating pipelines, a "resilience verification-driven" intelligent pump station collaborative regulation strategy was designed and dynamically implemented. Its core objective is not merely for daily energy saving or optimization, but rather to proactively create and utilize operating periods with different risk characteristics. By carefully constructing extreme operating states of "intelligent resource scarcity," the system's collaborative control capabilities are empirically tested in a graded and categorized manner. When the number of intelligent pump stations is large, the impact of external power outages is relatively small. Furthermore, when there are pump stations with a high frequency of power outages, the reliability of verifying the impact of these pump stations on the overall operating status under power outage conditions is likely to be high. Therefore, the collaborative regulation strategy is determined based on these factors.

[0082] This solution follows a closed-loop decision-making path of "risk profiling -> strategy matching -> scenario construction -> execution verification". The system continuously profiles its operational status in two ways: first, the internal intelligence level (the proportion of pump stations operating by intelligent agents); second, the external risk level (the power outage history and distribution of non-intelligent agent pump stations). Based on different risk combinations, different collaborative adjustment targets and verification intensities are dynamically matched. The core principle is that if a large number of intelligent agent pump stations are operating, the risk of power outages is relatively low, and intelligent agent resources should be proactively reduced to verify the system's inherent collaborative stability and minimum resource requirements in the absence of external disturbances. Conversely, if a high-risk node occurs, the system's preparedness and response capabilities for high-probability failures must be verified while maintaining a certain intelligent agent framework. The entire process transforms the operation scheduling itself into a continuous learning and verification experiment.

[0083] S31 uses the overlapping data between the intelligent pump station and the operating pump station to determine the operating pump station that belongs to the intelligent pump station and designates it as the intelligent operating pump station. Intelligent agent-operated pump stations: Among all currently operational pump stations, those that have deployed intelligent agents (possessing local computing, collaborative control, and emergency response capabilities). These are the core execution nodes for achieving advanced collaborative scheduling.

[0084] This step is fundamental for quantifying the "intelligence density" of the current operating system. Intelligent agent pump stations are the material basis for the system to achieve complex collaborative control and emergency response. By calculating the set of "intelligent agent operating pump stations," we can identify which nodes among the currently active pump stations possess "intelligence." This distinction is crucial because all subsequent decisions regarding collaborative feasibility and strategy strength are based on whether these intelligent nodes can form an effective control network. It provides the most direct input for assessing the system's current level of intelligence and collaborative potential.

[0085] It should be noted that the above steps include the following: S311 Based on the number of intelligent agent pump stations in the operating pump stations, determine whether the proportion of intelligent agent pump stations in the operating pump stations is greater than a preset proportion threshold. If yes, determine that collaborative scheduling can be performed; otherwise, proceed to step S312. The percentage of pump stations currently operating with intelligent agents: This is the percentage of pump stations currently operating with intelligent agents out of the total number of operating pump stations. It directly measures the "intelligent coverage rate" of the current operating shifts.

[0086] Preset ratio threshold: An empirical critical value used to determine whether the level of intelligence is sufficient to support complex global collaborative scheduling, such as 60%.

[0087] This step is a "fast track to global collaborative feasibility." When the proportion of pump stations operated by intelligent agents is sufficiently high (e.g., exceeding 60%), it means that the current operating system itself is already a highly intelligent network, and its impact from power outages is relatively low. Therefore, it can be directly determined that "collaborative scheduling can be performed," skipping the tedious checks on individual risk points and quickly entering the optimization calculation stage. This reflects the principle of decision-making efficiency, simplifying the process when conditions are favorable.

[0088] S312 determines whether there is an intelligent agent operating pump station among the operating pump stations. If yes, proceed to step S32; otherwise, determine that collaborative scheduling cannot be performed. This step, the "Minimum Resource Guarantee Check for Coordinated Scheduling," is the safety baseline of the decision-making process. The essence of coordinated scheduling is based on information exchange and joint decision-making among multiple intelligent nodes. If there isn't a single intelligent agent among all currently operating pump stations, there's no physical basis for implementing any coordinated strategy. In this situation, any attempt at coordination is futile, and the system must decisively determine that "coordinated scheduling cannot be performed" to prevent invalid or dangerous operations. This ensures that strategy generation only begins when the basic conditions are met.

[0089] S32 uses power outage data of pump stations other than intelligent body pump stations to determine the number of power outages of pump stations other than intelligent body pump stations. Pumping stations excluding smart agent pumping stations: All pumping stations in the pipeline network that have not deployed smart agents, i.e., "conventional pumping stations" or "non-smart pumping stations".

[0090] Number of power outages: The total number of power outages that occurred at each regular pumping station within a set statistical period (such as the past 12 months). This is a historical indicator of the power supply reliability of a single site.

[0091] This step aims to "quantify the historical performance of external risk sources." The safe implementation of the intelligent agent collaborative strategy highly depends on the understanding of the risk level of the non-intelligent components. By statistically analyzing the power outage history of conventional pumping stations, the system can construct a data profile of "external risks." This profile is used to determine whether the current external environment is generally stable or whether there are individual or widespread weaknesses. This judgment is the key basis for deciding what collaborative verification strategy to adopt subsequently. It realizes a shift in scheduling thinking from "blind optimism" to "risk perception."

[0092] It should be noted that the above steps include the following: S321 takes the pump stations other than the intelligent pump station as other pump stations, and determines whether there are other pump stations whose number of power outages exceeds the preset power outage threshold. If yes, proceed to step S322; otherwise, determine that collaborative scheduling can be performed. Preset power outage threshold: Define a frequency standard for a pumping station to be considered "high power outage risk", such as "more than 3 power outages per year".

[0093] This step is "high-risk individual screening." If no station among all regular pumping stations is found to have more than three power outages per year, it means that the overall historical performance of the non-intelligent components is very reliable, and external risks are at a low level. In this context, the probability of random sudden failures in the system is low, thus providing a large "safety margin" to try more challenging collaborative strategies, such as significantly reducing the number of online agents to verify the system's inherent resilience during "quiet periods." Therefore, it can be directly determined that "collaborative scheduling can be performed," and the optimization phase of constructing extreme scenarios can begin.

[0094] S322 identifies other pump stations with more than a preset power outage number threshold as power outage risk pump stations, determines whether there are any power outage risk pump stations among the operating pump stations, and if so, determines that collaborative scheduling can be performed; otherwise, proceeds to step S312. Pump stations at risk of power outage: These are conventional pump stations that have experienced more power outages than a preset threshold and are therefore marked as high-risk.

[0095] This step, handling the special case of "high-risk node online operation," is a highlight of the invention's logic. When a "power outage risk pump station" is detected and it is currently running, this effectively identifies a "high-probability fault verification window." Although external risks increase at this time, according to the core idea of ​​this embodiment, this is precisely an excellent opportunity for high-level, near-real-world verification. Because the probability of a power outage at this high-risk pump station is significantly higher than at other pump stations, the system has a higher chance of observing the actual fault impact and coordinated response process. Therefore, the system should not only not avoid it, but should seize the opportunity and determine that "coordinated scheduling processing can be performed." The aim is to verify the system's isolation and compensation capabilities in the face of high-probability single-point failures under controlled conditions.

[0096] S33 determines the coordinated adjustment strategy for the intelligent agent pumping stations based on the data from the operation of the pumping stations by the intelligent agent and the number of power outages of pumping stations other than the intelligent agent pumping stations. Furthermore, the proportion of the power outage risk pump station among the other pump stations is determined based on the number of power outages. It is then determined whether the proportion of the power outage risk pump station among the other pump stations is less than a preset power outage risk pump station proportion threshold. If so, it is determined that collaborative scheduling can be performed; otherwise, it is determined that collaborative scheduling cannot be performed.

[0097] Percentage of pump stations at risk of power outage: The number of pump stations marked as "risk of power outage" out of the total number of "other pump stations" (conventional pump stations).

[0098] Preset power outage risk threshold for pump stations: a critical value for judging whether the risk is "common", such as 25%.

[0099] This step is the "final determination of the overall risk level" and the last safety gate before initiating active verification. Its significance lies in whether the operation of the pump station can be effectively verified.

[0100] If the proportion of risky pump stations is very low (e.g., <25%), it indicates that the unstable factors are only a very small number of cases, and the non-intelligent parts of the system are difficult to verify effectively. This allows for collaborative scheduling to determine the impact of pump stations set up by non-intelligent entities when a power outage occurs.

[0101] If the proportion of risky pump stations is very high (≥25%), it means that the overall network reliability of other pump stations is very poor and their power outage frequency is high. At this time, the probability of simulating other pump stations on the long-distance pipeline network is high. If there are not enough pump stations in operation by the intelligent agent, it is determined that "cooperative scheduling processing cannot be carried out", that is, the current operating state is maintained.

[0102] It should be noted that when collaborative scheduling is possible, the constraints are that at least a preset proportion of intelligent agent pump stations are in operation and that the user's heating demand can be met. The goal is to minimize the number of intelligent agent pump stations in operation. This ensures that, under the condition of intelligent agent pump stations in operation, the operational demand can be met in the event of a power outage.

[0103] Specific example of coherence: Suppose a long-distance heating pipeline network has 12 pumping stations, of which 5 have deployed intelligent systems (S1-S5) and 7 are conventional pumping stations (C1-C7). Currently, 8 pumping stations need to be activated to meet the heating demand.

[0104] S31 execution: The currently operating pump stations are [S1, S2, S4, C2, C3, C5, C6, C7]. Among them, S1, S2, and S4 are pump stations operated by the intelligent agent (3 stations).

[0105] S311 execution: The proportion of the pump station in operation by the intelligent agent = 3 / 8 = 37.5%, which is less than the preset threshold of 60%, proceed to S312.

[0106] S312 execution: If there is an intelligent agent operating pump station (S1, S2, S4) in the operating pump station, proceed to S32.

[0107] S32 execution: Historical data shows that conventional pumping stations C5 experienced 5 power outages and C6 experienced 4 power outages in the past year, both exceeding the preset threshold (3 times). Other conventional pumping stations experienced ≤1 power outage.

[0108] S321: Pump stations (C5, C6) with more than the threshold number of power outages are identified; proceed to S322.

[0109] S322: Both C5 and C6 are marked as power outage risk pump stations, and they are currently in the operating queue. Therefore, the current period is determined to be a "high-probability fault verification window," and coordinated scheduling can be performed.

[0110] S33 execution: Calculate the percentage of risky pumping stations: Among the 7 conventional pumping stations, 2 (C5, C6) are risky pumping stations, accounting for approximately 28.6%. Assuming the preset percentage threshold is 25%, 28.6% > 25%, and according to S33, it is determined as "not allowed". However, note that in this embodiment, S322 has already determined as "allowed" based on the strong feature of "high-risk pumping stations online", so it will not be executed here in actual circumstances.

[0111] The system enters "high-risk verification mode". Assume the minimum preset agent operation rate is 35% for the number of pump stations (this rate is increased due to high-risk points). Then, at least... = 1.05 ≈ 1 intelligent body pump station online.

[0112] Optimization calculation: Constrained by meeting heating requirements and ensuring at least one agent is online, the objective is to minimize the number of operating agents. After hydraulic model calculations, the optimal solution is determined: keep S1, S4, and S5 online, put S2 into standby mode, and start conventional pump station C1 for supplementary power.

[0113] Generate collaborative verification run plan: New run combination: [S1, S4, S5, C1, C2, C3, C5 (high risk), C6 (high risk)].

[0114] Status: The intelligent agent is operating three pump stations (S1, S4, S5), meeting the minimum framework requirement (3 stations). The system has entered the "preparatory impact resistance" verification state. The intelligent agents S1, S4, and S5 have been optimized in layout to form the best monitoring and compensation posture for C5 and C6.

[0115] Verification and Observation: The system operates in this state. Due to the high probability of failure of C5 and C6, the system is continuously monitored to capture possible real power outage events and record the coordinated response process of S1, S4, and S5, thereby verifying the system's shock resistance under extreme pressure.

[0116] It should be noted that the coordinated adjustment combination of the intelligent body pump station is a combination of equipment that was in operation in the past when the intelligent body pump station was in a stopped state.

[0117] Furthermore, the method for determining the control and management method of the setting target of the intelligent agent is as follows: Establish and implement a "data-driven intelligent agent deployment verification and dynamic management" decision-making system. Its core objective is to utilize the actual performance data of "coordinated adjustment combinations" (i.e., alternative operation schemes when intelligent agent pump stations are out of service) formed during historical operation to assess the overall reliability of existing intelligent agent pump stations. Based on this assessment result and the actual failure impact of non-intelligent pump stations, intelligent decisions are made regarding whether and how to adjust subsequent intelligent agent deployment (additional) plans. This constitutes a complete closed loop from "operational verification" to "investment decision-making," ensuring that every additional investment in intelligent agents is based on empirically tested system shortcomings.

[0118] This method follows a logical chain of "backtracking verification -> reliability profiling -> impact assessment -> target decision-making". The system first extracts all "cooperative adjustment combinations" and their corresponding "cooperative runtime phases" (S41) from historical data, which is equivalent to a complete historical record of "stress tests" in scenarios where agents are absent. Next, the system analyzes the "attendance rate" and stability of each agent pump station in all relevant tests to screen out "reliable pump stations" (S42), i.e., those agents that have been fully validated under various challenging scenarios while in a shutdown state. Then, the system assesses the risk of non-intelligent pump stations by combining the actual impact of power outages during these "stress tests" (especially the extent to which they extend system adjustment time) (S43). Finally, by combining the "reliability saturation of deployed agents" and the "risk exposure of undeployed pump stations", the system decides whether to maintain the existing deployment strategy and continue validation, or to list specific high-impact pump stations as new deployment targets.

[0119] S41 determines the coordinated operation period under the coordinated regulation combination based on the operation results of the intelligent pump station. Coordinated regulation combination: refers to the alternative operating equipment combination (including other intelligent agent pump stations and conventional pump stations) adopted by the system to maintain heating when one or more intelligent agent pump stations are out of service during historical operation. It records the specific scheduling scheme under a "temporary shortage of intelligent agent resources" event.

[0120] Coordinated operation period: refers to the continuous time period during which the above-mentioned "coordinated adjustment combination" is executed and operates stably in actual production. It records how long the combination "survives" under real working conditions and is direct evidence for evaluating the effectiveness of the combination.

[0121] This step is crucial for "building an empirical analysis database." The "coordinated adjustment combinations" are not theoretical deductions, but rather contingency plans actually used by the system in past operations to address agent absence. Extracting these combinations and their corresponding stable operating periods is equivalent to collecting all historical reports of the system's "natural stress tests" under different agent absence scenarios, reflecting the operational reliability of the long-distance pipeline network. This provides irreplaceable raw material for subsequent assessments of individual agent value and risks of conventional pumping stations.

[0122] S42 determines the identified reliable pump stations among the intelligent pump stations based on the cooperative operation segments under different combinations of cooperative adjustment; Identifying reliable pump stations: In historical data analysis, for each "coordinated adjustment combination" that includes the shutdown of the intelligent agent pump station, the corresponding "coordinated running period" exceeds a preset running time threshold. This means that whenever the system needs to "bypass" the operation of this pump station, it is reliably verified within a sufficiently long time.

[0123] Preset runtime segment quantity threshold: A time standard for judging whether an alternative is "sufficiently stable", such as "cooperative runtime segments need to exceed 3 days".

[0124] This step aims to "reverse define the reliability of an agent." If all "absent scenarios" (coordinated regulation combinations) of an agent pump station have been tested and verified over a sufficiently long period of time, its operational reliability in the long-distance pipeline network under absent conditions has been reliably verified, and therefore it is used to identify a reliable pump station.

[0125] S43 determines the control and management method for the setting target of the intelligent agent based on the identified reliable pump station data and the power outage impact data of pump stations other than the intelligent agent pump station during the cooperative operation period.

[0126] It should be noted that the coordinated operation period refers to the time period during which the pumping station operates under a coordinated regulation combination.

[0127] It is understood that the identified reliable pumping station is an intelligent pumping station whose number of operating segments under all coordinated adjustment combinations is greater than a preset threshold for the number of operating segments.

[0128] It should be noted that the management and control method for determining the setting target of the intelligent agent, based on the identified reliable pump station data and the power outage impact data of pump stations other than the intelligent agent pump station, specifically includes: S431 takes the pump stations other than the intelligent pump station as other pump stations, and determines the collaborative adjustment duration of the other pump stations during the power outage period in the collaborative operation period based on the power outage impact data of the other pump stations. It then determines whether there is a power outage period with a collaborative adjustment duration longer than the preset collaborative adjustment duration threshold. If yes, proceed to step S432; otherwise, proceed to step S433. Preset coordination duration threshold: a time standard used to determine whether the impact is "severe", such as "coordination duration exceeds 30 minutes".

[0129] This step is the "screening for severe risk events." During stress testing (cooperative operation phase), if a power outage at another pumping station causes the system to enter a prolonged period of adjustment and instability (e.g., exceeding 30 minutes), it indicates that the existing intelligent pumping stations may not be able to meet the reliable operation requirements of long-distance pipelines, and a power outage at another pumping station would greatly exacerbate the crisis. Identifying these "severely impactful" power outage events is crucial for pinpointing the highest priority deployment targets.

[0130] S432 takes the power outage period with a coordinated adjustment duration longer than the preset coordinated adjustment duration threshold as the affected period, and determines whether the number of the affected periods is greater than the preset affected period number threshold. If so, it determines that the control and processing method of the setting target of the intelligent agent is that there are other pump stations with a coordinated adjustment duration longer than the preset coordinated adjustment duration threshold in the coordinated adjustment period. If not, proceed to step S433. Threshold for the number of impact periods: a standard for judging whether a risk is "frequent" rather than "occasional", such as "this type of serious impact period occurs more than 5 times".

[0131] This step represents an upgrade from "single event to pattern recognition." A single severe event may be accidental, but if it recurs repeatedly in history (exceeding a threshold), it indicates that other pumping stations, under their current operating conditions, may be causing unreliable operation of the long-distance pipeline network. Therefore, the system decision is very clear: these other pumping stations that have caused long-term adjustment difficulties in historical pressure tests are listed as new targets for the intelligent agent's settings.

[0132] S433 determines whether all intelligent agent pump stations belong to the category of reliable pump stations. If so, it determines that no new intelligent agent settings are required. If not, it continues to perform collaborative adjustment processing according to the collaborative adjustment strategy of the intelligent agent pump stations until the preset time or all intelligent agent pump stations belong to the category of reliable pump stations.

[0133] It should be noted that when the preset duration is not reached or not all intelligent pump stations belong to the identified reliable pump stations, if the number of the affected time periods is greater than the preset threshold for the number of affected time periods, the control and processing method for determining the setting target of the intelligent agent is to find other pump stations with a collaborative adjustment duration greater than the preset threshold for collaborative adjustment duration during the collaborative adjustment period.

[0134] Preset duration: A total verification period, such as "continuous verification for 6 months".

[0135] This step, "Deployment Saturation and Verification Completeness Check," represents another path for control decision-making. If the analysis reveals that all existing agent pump stations have been verified as "reliable pump stations," it indicates that the existing agent network has reached a high level of "reliability saturation"—meaning that from a system resilience perspective, the existing deployment is robust enough, and the operational reliability of the heating network with different agent pump stations in shutdown status has been fully verified. Simultaneously, if there are no frequent severe impact events (S432 not triggered), it indicates that the risks of unprotected portions are relatively controllable. In this case, the system can confidently determine that "no new agent configuration is needed," suspending further capital investment, signifying that the deployment goals for the current stage have been successfully achieved. If this state is not reached, the system instructions continue to execute the coordinated adjustment strategy to generate more verification data, driving the system towards this complete state.

[0136] Suppose a heating network has 8 pumping stations, of which 4 are intelligent pumping stations (A1, A2, A3, A4) and 4 are conventional pumping stations (B1, B2, B3, B4). The system has been running for two years.

[0137] S41 Extract historical data: Analyze historical logs to identify all operational records during the shutdown of the intelligent pump station. For example, it was found that when A1 was shut down, the system repeatedly used the combination [A2, A3, A4, B1, B2] to operate, and each time it ran stably for more than 8 hours (meeting the >4-hour threshold). Similarly, identify various alternative combinations and their durations when A2, A3, and A4 were shut down.

[0138] S42 Reliability Assessment: For A1: All combinations excluding A1 have a runtime >3 days. → A1 is marked as "Identifying a reliable pumping station".

[0139] For A2: All combinations excluding A2 have a runtime >3 days. → A2 is marked as "Identifying a reliable pumping station".

[0140] For A3: Most combinations ran for more than 3 days, but one combination [A1, A4, B1, B3, B4] only stabilized for 2 hours. → A3 was not marked as an identified reliable pumping station.

[0141] For A4: All combinations excluding A4 have a runtime >3 days. → A4 is marked as "Identifying a reliable pumping station".

[0142] S43 Comprehensive Decision-Making: S431 Screening for Critical Events: Retrieve regular pump station power outage records across all "cooperative operation periods." It was found that pump station B2 experienced three power outages, occurring within the A1 shutdown combination [A2, A3, A4, B1, B2], resulting in system adjustment and recovery times of 50, 45, and 55 minutes respectively (all > 30-minute threshold).

[0143] S432 pattern recognition: All pump stations caused 3 "impact periods". The preset threshold for the number of impact periods is 5. Since 3 is not greater than 5, proceed to the next step; S433 Completeness Check: Not all intelligent agent pump stations are “identified reliable pump stations” (A3 is not).

[0144] Final decision: The control and management method for setting the target of the intelligent agent is determined to be: continue to verify until 6 months or A3 is marked as a reliable pumping station.

[0145] This embodiment constructs a rigorous closed-loop management system for the entire lifecycle of intelligent agents, based on "operation as verification and data-driven decision-making," and establishes an evidence-driven investment decision-making mechanism. It completely changes the current situation where equipment upgrades rely on subjective experience or simple rules, and provides an objective answer to the investment question of "which pump station to install an intelligent agent for," based on a large amount of historical operating data. This significantly improves the scientific nature of capital expenditure and the rate of return on investment. It innovatively defines and measures the reliability value of an intelligent agent by "the system's performance when the intelligent agent is absent." This is a deeper evaluation perspective oriented towards the overall resilience of the system, promoting a shift in thinking from focusing on single-point availability to focusing on network resilience.

[0146] Example 2 In a second aspect, the present invention provides a computer system comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-described method for the operation and management of a long-distance pipeline network based on an intelligent agent when running the computer program.

[0147] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0148] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0149] The above description is merely one or more embodiments of this specification and is not intended to limit this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of this specification.

Claims

1. A method for operation and management of long-distance pipeline networks based on intelligent agents, characterized in that, Specifically, it includes: Using the operational data of the long-distance pipeline network, the operating pump stations of the long-distance pipeline network in different time periods are identified. Based on the operational pump station data in different time periods, the setting strategy of the intelligent agent based on the power outage data of the pump station is determined. The setting strategy is used to set up the intelligent agent. Based on the distribution data of the intelligent agent and combined with the matching degree of the pump station with intelligent agent in the power outage event, when it is determined that the setting target of the intelligent agent needs to be controlled, the next step is taken. Pump stations with intelligent agents are designated as intelligent agent pump stations. Based on the overlapping data between the intelligent agent pump stations and the operating pump stations, and excluding the power outage data of pump stations other than the intelligent agent pump stations, a collaborative adjustment strategy for the intelligent agent pump stations is determined. Based on the collaborative adjustment strategy, the collaborative adjustment processing of the intelligent agent pump stations is performed to obtain the operating results. Based on the operational results under the coordinated adjustment combination of different intelligent agent pump stations, the control and processing methods for the setting objectives of the intelligent agents are determined.

2. The operation and management method for long-distance pipeline networks based on intelligent agents as described in claim 1, characterized in that, The operating pump stations during the specified time period are those that are in operation during that time period.

3. The operation and management method for long-distance pipeline networks based on intelligent agents as described in claim 1, characterized in that, The method for determining the setting strategy of the intelligent agent based on power outage data in the pumping station is as follows: The percentage of pump stations in operation during different time periods is determined by using data on operating pump stations during those different time periods. Based on the stated proportions, determine the time periods within the stated time period that require reliable operation. Based on the distribution data of the reliable operation demand period, the setting strategy of the intelligent agent of the pumping station based on power outage data is determined.

4. The operation and management method for long-distance pipeline networks based on intelligent agents as described in claim 3, characterized in that, The reliable operation requirement period is the period in which the proportion of the quantity is greater than a preset proportion threshold.

5. The operation and management method for long-distance pipeline networks based on intelligent agents as described in claim 3, characterized in that, Based on the distribution data of the reliable operating demand period, the setting strategy of the intelligent agent of the pumping station based on power outage data is determined, specifically including: If the proportion of reliable operation demand periods in the time period is greater than the preset time period quantity proportion threshold, then the setting strategy of the intelligent agent of the pump station based on power outage data is determined to be to set the pump station with power outage duration greater than the preset duration threshold as the intelligent agent setting target. If the proportion of reliable operation demand periods in the time period is not greater than the preset time period quantity proportion threshold, then the setting strategy of the intelligent agent of the pump station based on power outage data is determined to be to set the pump station with power outage duration greater than the second preset duration threshold as the intelligent agent setting target.

6. The operation and management method for long-distance pipeline networks based on intelligent agents as described in claim 1, characterized in that, The distribution data of the agents includes the number of pump stations where agents exist.

7. The operation and management method for long-distance pipeline networks based on intelligent agents as described in claim 1, characterized in that, Determine the control and management processes required for setting up intelligent agents, specifically including: Based on the distribution data of the intelligent agents, the pump stations where intelligent agents exist are identified, and the pump stations where intelligent agents exist are designated as intelligent agent pump stations. Based on the matching degree of the intelligent pump station in the power outage event, determine the pump stations that do not belong to the intelligent pump station in the power outage event, and take them as the affected pump stations. Based on the data from the intelligent pump station and the impact of power outage events on the pump station, it is determined whether the intelligent agent's target setting management process is required.

8. The operation and management method for long-distance pipeline networks based on intelligent agents as described in claim 7, characterized in that, In the above steps, the influence coefficient is determined based on the proportion of intelligent agent pump stations in the pump station and the proportion of the number of power outage events in the power outage events. When the influence coefficient is greater than the preset influence coefficient threshold, it is determined that no control processing of the intelligent agent setting target is required.

9. The operation and management method for long-distance pipeline networks based on intelligent agents as described in claim 1, characterized in that, The method for determining the control and processing method of the setting target of the intelligent agent is as follows: Based on the operational results of the coordinated regulation combination of intelligent pump stations, determine the coordinated operation period under the coordinated regulation combination; Based on the cooperative operating segments under different combinations of cooperative adjustment, the identified reliable pump stations in the intelligent pump station are determined. Based on the identified reliable pump station data and the power outage impact data of pump stations other than the intelligent agent pump station during the collaborative operation period, a management and control method for the setting target of the intelligent agent is determined.

10. A computer system, comprising: A memory and processor connected by communication, and a computer program stored in the memory and capable of running on the processor, characterized in that, when the processor runs the computer program, it executes an operation management method for a long-distance pipeline network based on an intelligent agent as described in any one of claims 1-9.

Images