Energy access method and system based on offshore risk construction

Abstract

The invention provides an energy access method and system based on offshore risk construction, and the method comprises the steps: collecting data sources, and constructing a risk assessment model for each data source for risk assessment; carrying out weighted fusion on the risk assessment value of each data source; designing a self-adaptive scheduling decision mechanism, and dynamically adjusting an energy access strategy according to the overall risk assessment value; introducing closed-loop feedback control and scheduling strategy optimization to ensure that the system can perform dynamic adjustment according to real-time feedback of a scheduling decision in actual operation to obtain an optimized scheduling decision; and carrying out comprehensive evaluation on the optimized scheduling decision and a plurality of evaluation indexes, and providing intelligent decision support for a decision maker based on the comprehensive evaluation. The intelligent level of offshore energy access is improved, and a feasible technical path is provided for future offshore energy system construction.

Description

Technical Field

[0001] The present invention belongs to the field of offshore energy dispatching, and in particular relates to an energy access method and system based on offshore risk construction. Background Art

[0002] With the continuous growth of offshore energy demand, energy forms such as offshore wind power, tidal energy, and wave energy have become an important part of the global energy structure transformation. However, the construction and operation of offshore energy access systems still face many challenges. Offshore energy access systems usually rely on hardware facilities such as submarine cables and offshore power generation platforms to transmit energy generated at sea to the land energy network. This process involves a large number of equipment, communication systems, and scheduling strategies, and the complex and dynamic natural environment at sea puts the reliability and stability of the system to a severe test. The main risk factors include but are not limited to severe weather at sea (such as storms, typhoons, ocean currents, etc.), equipment failure, system overload, grid imbalance, etc. These factors will affect the stable access of energy and even lead to energy supply interruptions or system failures.

[0003] Currently, offshore energy access systems designed to address these issues primarily rely on fixed, pre-set scheduling strategies and traditional fault-tolerance mechanisms. Most existing systems rely on predictive models and fixed fault-tolerance mechanisms to address known risks, such as using meteorological data to predict storms and prepare for equipment shutdowns in advance. However, this approach has two significant drawbacks: First, fixed, pre-set strategies lack flexibility and real-time responsiveness, making them incapable of adapting to sudden or unexpected environmental changes. Second, traditional fault-tolerance mechanisms focus more on emergency response, often addressing problems only after they occur, rather than effectively preventing potential risks at their source. For example, many systems only shut down when meteorological data indicates an impending typhoon, by which time the system may have already been affected to some extent, and recovery often takes a long time. Furthermore, traditional offshore energy access systems lack multi-layered awareness and dynamic response mechanisms for heterogeneous risks, making them unable to effectively classify, prioritize, and implement appropriate response strategies for various risk factors. The systems also suffer from relatively low levels of flexibility and intelligence, making it difficult to optimize and dynamically adjust energy access strategies based on real-time data.

[0004] More critically, the existing system fails to make timely policy adjustments based on real-time data when encountering different types of risks. Whether it's a natural disaster or equipment failure, the existing system relies heavily on pre-set rules and policies to respond. This makes it difficult to react quickly to sudden changes, increasing the risk of energy supply disruptions. Furthermore, when multiple risks occur simultaneously, the existing offshore energy access system fails to effectively implement risk stratification and optimized scheduling. It is unable to optimize access strategies in an environment with multiple risks, avoid resource waste, and ensure overall system stability. Summary of the Invention

[0005] The purpose of this invention is to propose an energy access method and system based on offshore risk management. By combining adaptive risk response, heterogeneous risk perception and hierarchical assessment, and a closed-loop feedback control mechanism, this method effectively addresses the existing system's shortcomings of poor flexibility, slow response, and inability to cope with complex and changing risk environments. This significantly improves the security, stability, and economic efficiency of offshore energy access systems. This innovative solution not only enhances the intelligence level of offshore energy access but also provides a viable technical path for future offshore energy system construction.

[0006] To achieve the above objectives, a first aspect of the present invention provides an energy access method based on offshore risk construction, the method comprising the following steps:

[0007] S1. Collect data sources, build a risk assessment model for each data source to perform risk assessment, and obtain a risk assessment value for each data source; wherein the data sources include meteorological data, ocean data, and equipment data;

[0008] S2. Perform weighted fusion on the risk assessment values ​​of each data source to obtain an overall risk assessment value. Based on the overall risk assessment value, determine whether an emergency response is required:

[0009] If the overall risk assessment value exceeds the set risk threshold, the system's emergency response mechanism will be triggered to ensure stable system operation;

[0010] S3. Build a real-time feedback mechanism to optimize the risk assessment model based on actual operation results;

[0011] S4. Design an adaptive scheduling decision mechanism to dynamically adjust the energy access strategy based on the overall risk assessment value and calculate the final scheduling decision, including:

[0012] Design the initial adaptive scheduling decision function to adjust the system's energy access or load scheduling strategy according to the overall risk assessment value:

[0013] When the overall risk assessment value is less than the preset risk threshold, it means that the system is in a low-risk state and energy access is flexible;

[0014] When the overall risk assessment value is greater than or equal to the preset risk threshold, it means that the system is in a high-risk state and energy access needs to be restricted;

[0015] Design an adaptive feedback mechanism to dynamically adjust the adjustment coefficient of the scheduling sensitivity of the initial adaptive scheduling decision function and the adjustment amount of the scheduling benchmark value based on the actual operation effect after each scheduling;

[0016] The initial adaptive scheduling decision function after dynamic adjustment is modified by combining external risk factors to obtain the scheduling decision function, which is expressed as:

[0017] D(R total ,R ext )=(λ1R total +λ2)·w ext

[0018] Among them, D(R total ,R ext ) is the scheduling decision, which indicates the change in energy access. A negative value indicates a decrease in access, while a positive value indicates an increase in access. R total is the overall risk assessment value, R ext is the external environmental risk factor, w ext is the weighted coefficient of external environmental risk, λ1 is the adjustment coefficient of scheduling sensitivity, and λ2 is the adjustment amount of the scheduling benchmark value;

[0019] S5. By introducing closed-loop feedback control and scheduling strategy optimization, we ensure that the system can dynamically adjust according to the real-time feedback of scheduling decisions during actual operation to obtain optimized scheduling decisions;

[0020] S6. Comprehensively evaluate the optimized scheduling decision with multiple evaluation indicators, and provide intelligent decision support to decision makers based on this.

[0021] Furthermore, the risk assessment model includes a meteorological data risk assessment model, an ocean data risk assessment model, and an equipment data risk assessment model, which are specifically expressed as follows:

[0022]

[0023] Among them, α1 is a hyperparameter related to the sensitivity of meteorological data fluctuations, std(X1) is the standard deviation of meteorological data X1, α2 is a hyperparameter related to the sensitivity of marine environment fluctuations, std(X2) is the standard deviation of marine data X2, α3 is a hyperparameter related to the equipment failure history, std(X3) is the standard deviation of equipment data X3.

[0024] Furthermore, the risk assessment value of each data source is weighted and integrated to obtain an overall risk assessment value, specifically including:

[0025] Based on the historical accuracy and real-time volatility of each data source, weighting factors w1, w2, and w3 are calculated for each data source to dynamically adjust the contribution of each data source to the overall risk assessment value. The calculation is as follows:

[0026]

[0027] Among them, σ1, σ2, and σ3 are the historical standard deviations of meteorological data, ocean data, and equipment data, respectively, indicating the volatility of each data source; w1, w2, and w3 are the dynamic weighting coefficients of data sources X1, X2, and X3, respectively, indicating the weight of the risk assessment of each data source in the overall assessment;

[0028] The risk assessment value R(X i ) and the corresponding weight w i Perform weighted summation to obtain the overall risk assessment value R total , the formula is as follows:

[0029] R total =w1·R(X1)+w2·R(X2)+w3·R(X3)

[0030] Among them, R total It is the overall risk assessment value, reflecting the overall risk level of the system under current conditions.

[0031] Furthermore, the real-time feedback mechanism is constructed to optimize the risk assessment model based on actual operation results, specifically including:

[0032] After the system makes an energy access decision, it collects various data in real time and feeds this data into various risk assessment models. By using this feedback data, the hyperparameters in the model are adjusted, thereby optimizing the risk assessment process.

[0033] Based on real-time feedback data, the risk assessment model is retrained to reflect changes in the current environment.

[0034] Furthermore, the adaptive feedback mechanism specifically includes:

[0035] Monitor the real-time results after scheduling and generate feedback signals. If the feedback signals indicate that the scheduling effect is not ideal, adjust the adjustment coefficient of the scheduling sensitivity and the adjustment amount of the scheduling reference value, which can be expressed as:

[0036] λ1=λ1+Δλ1and λ2=λ2+Δλ2

[0037] Among them, Δλ1 and Δλ2 are adjustment amounts calculated based on the feedback signal. λ1 is the adjustment coefficient of the scheduling sensitivity, which is dynamically adjusted according to the feedback results; λ2 is the adjustment amount of the scheduling benchmark value, which corrects the benchmark scheduling amount in real time.

[0038] Furthermore, the S5 specifically includes:

[0039] The feedback factor is introduced to modify the initial scheduling decision based on real-time feedback to ensure that the system can respond to external disturbances in a timely manner. The adjusted scheduling decision function is:

[0040] D opt (R total ,R ext ,β)=D(R total ,R ext )·(1+β)

[0041] Among them, D opt (R total ,R ext ,β) is the adjusted scheduling decision, which adjusts the initial decision according to the feedback factor β;

[0042] Define the error function E, which represents the gap between the actual scheduling behavior and the target;

[0043] Define an update rule to adjust β according to the error function E, and adjust the scheduling strategy according to the impact of the error on the feedback factor; wherein the update rule is expressed as:

[0044]

[0045] Where α is the error feedback sensitivity coefficient, which is used to adjust the direct impact of the error E on β; is the rate of change of the error, reflecting the trend of the error over time; λ is the time response coefficient, which is used to adjust the dynamic response degree of the error change to β;

[0046] Combining the error E and the feedback factor β, a multi-objective optimization function is designed to comprehensively evaluate and optimize the scheduling decision. opt ), expressed as:

[0047] O(D opt )=w1·E+w2·L+w3·S+w4·C

[0048] Among them, w1, w2, w3, and w4 are weight coefficients used to balance the priorities of various objectives; E is the error, which indicates the deviation between the actual and expected; L is the load, which indicates the current load of the equipment; S is the system stability, which reflects the stability level of the system under the current scheduling; C is the energy efficiency, which indicates the efficiency of energy utilization.

[0049] Furthermore, in the multi-objective optimization function O(D opt ) adds a regularization term R to control the volatility of scheduling decisions and limit the range of changes in scheduling decisions, which can be expressed as:

[0050] O regularized (D opt )=O(D opt )+γ·R(D opt )

[0051] Among them, O regularized (D opt ) is the regularized multi-objective optimization function, R(D opt ) is the change in scheduling decision, which is calculated as: R(D opt )=|D opt (t)-D opt (t-1)|, which represents the difference between the current scheduling decision and the scheduling decision at the previous moment; γ is the regularization coefficient, which controls the magnitude of the fluctuation of the scheduling decision.

[0052] Furthermore, the S6 specifically includes:

[0053] An evaluation model based on multi-objective optimization is constructed to obtain a comprehensive evaluation score; wherein the evaluation model based on multi-objective optimization is calculated as follows:

[0054]

[0055] Among them, Q(D opt ) is the comprehensive evaluation score, which represents the comprehensive effect of the scheduling decision; w i is the objective function f i (D opt ) is the weight coefficient of the total evaluation score; f i (D opt ) is the evaluation index function, which represents the contribution of the i-th evaluation index to the scheduling decision;

[0056] Multiple evaluation metrics were designed, covering different key performance areas, including but not limited to system load, energy consumption, operating cost, and stability;

[0057] Provide support to decision makers based on comprehensive assessment scores, including:

[0058] Based on the current comprehensive evaluation score, simulate future operating scenarios under different decisions and generate predictions of possible scheduling strategies;

[0059] Based on historical data, comprehensive evaluation scores, and scenario simulation results, machine learning models are used to intelligently recommend optimal scheduling decisions.

[0060] The evaluation scores and recommended strategies of different scheduling decisions are displayed through a visual interface.

[0061] Furthermore, according to the uncertainty in the actual operating environment, a risk assessment function R(D opt ), which is used to consider the impact of these uncertainties on scheduling decisions, is expressed as:

[0062] R(D opt )=E[f(D opt )]

[0063] Among them, R(D opt ) is the risk assessment score, which indicates the impact of uncertainty on scheduling decisions; f(D opt ) is the evaluation function, which represents the risk level under different scheduling decisions; E[f(D opt )] represents the expected value of the evaluation function, and the weighted average value under all possible risk scenarios is calculated.

[0064] In a second aspect of the present invention, there is provided an energy access system based on offshore risk construction, the system comprising:

[0065] A data source acquisition unit is used to acquire data sources, construct a risk assessment model for each data source to perform risk assessment, and obtain a risk assessment value for each data source; wherein the data sources include meteorological data, ocean data, and equipment data;

[0066] The risk assessment unit is used to perform weighted fusion on the risk assessment values ​​of each data source to obtain an overall risk assessment value. Based on the overall risk assessment value, it is determined whether an emergency response is required:

[0067] If the overall risk assessment value exceeds the set risk threshold, the system's emergency response mechanism will be triggered to ensure stable system operation;

[0068] Evaluation and optimization unit, used to build a real-time feedback mechanism and optimize the risk assessment model based on actual operation results;

[0069] The energy access analysis unit is used to design an adaptive scheduling decision mechanism, dynamically adjust the energy access strategy according to the overall risk assessment value, and calculate the final scheduling decision, including:

[0070] Design the initial adaptive scheduling decision function to adjust the system's energy access or load scheduling strategy according to the overall risk assessment value:

[0071] When the overall risk assessment value is less than the preset risk threshold, it means that the system is in a low-risk state and energy access is flexible;

[0072] When the overall risk assessment value is greater than or equal to the preset risk threshold, it means that the system is in a high-risk state and energy access needs to be restricted;

[0073] Design an adaptive feedback mechanism to dynamically adjust the adjustment coefficient of the scheduling sensitivity of the initial adaptive scheduling decision function and the adjustment amount of the scheduling benchmark value based on the actual operation effect after each scheduling;

[0074] The initial adaptive scheduling decision function after dynamic adjustment is modified by combining external risk factors to obtain the scheduling decision function, which is expressed as:

[0075] D(R total ,R ext )=(λ1R total +λ2)·w ext

[0076] Among them, D(R total ,R ext ) is the scheduling decision, which indicates the change in energy access. A negative value indicates a decrease in access, while a positive value indicates an increase in access. R total is the overall risk assessment value, R ext is the external environmental risk factor, w ext is the weighted coefficient of external environmental risk, λ1 is the adjustment coefficient of scheduling sensitivity, and λ2 is the adjustment amount of the scheduling benchmark value;

[0077] The access optimization unit is used to introduce closed-loop feedback control and scheduling strategy optimization to ensure that the system can dynamically adjust according to the real-time feedback of scheduling decisions during actual operation and obtain optimized scheduling decisions;

[0078] Personalized support unit, used to comprehensively evaluate the optimized scheduling decision with multiple evaluation indicators, and provide intelligent decision support to decision makers based on this

[0079] The beneficial technical effects of the present invention are at least as follows:

[0080] This paper addresses the shortcomings of existing technologies by proposing an offshore energy access system based on adaptive risk response and hierarchical perception. By integrating real-time data perception, hierarchical risk assessment, closed-loop feedback control, and intelligent optimization and scheduling, this system achieves real-time perception and dynamic response to multiple offshore risks, significantly improving the system's flexibility, intelligence, and security, and addressing several pain points in existing systems.

[0081] First of all, one of the core innovations of the present invention is the heterogeneous risk layered perception mechanism. Existing systems usually rely on a single type of data source or a preset risk prediction model, while the present invention divides offshore risks into physical risk layer, equipment risk layer and system risk layer by introducing multi-level data fusion technology. Each layer of risk is monitored and analyzed in real time through different data sources, so as to more accurately identify and evaluate different types of risks. By perceiving these risks in a hierarchical and heterogeneous manner, the present invention can formulate more detailed and flexible response strategies for different types of risks. For example, when facing natural disasters (such as storms and typhoons), the system will give priority to adjusting the load of energy access points; when facing equipment failures or grid overloads, the system will automatically start backup equipment or adjust the flow of power.

[0082] Secondly, the present invention introduces an adaptive risk response mechanism to solve the problem of insufficient flexibility of fixed preset scheduling strategies in the prior art. In the present invention, the system dynamically adjusts the energy access strategy according to environmental changes and real-time results of risk assessment through real-time feedback. By applying reinforcement learning algorithms or multi-objective optimization algorithms, the system can respond promptly to sudden risks and automatically adjust the start and stop of access points, load distribution, etc., to ensure that energy access can operate efficiently and stably in complex offshore environments. Unlike traditional fault-tolerant mechanisms, the system can not only respond to emergencies after a problem occurs, but also deploy in advance through prediction and prejudgment mechanisms before risks occur, thereby reducing the risk of energy supply interruptions.

[0083] Finally, this invention also proposes an optimized scheduling method based on closed-loop feedback control. Through this closed-loop control mechanism, the system continuously collects environmental and device data to adjust and optimize access policies in real time. This mechanism not only enables the system to self-regulate in response to different risk events, but also enables it to adopt the optimal resource scheduling strategy for the concurrent occurrence of multiple risks, thereby avoiding resource waste and improving energy supply efficiency. For example, under the dual impacts of severe weather and equipment failure, the system can intelligently dispatch power resources, optimizing system performance and reducing economic losses while ensuring safety.

[0084] In summary, the offshore energy access system proposed in this invention effectively addresses existing systems' shortcomings of poor flexibility, slow response, and inability to cope with complex and changing risk environments through the integration of adaptive risk response, heterogeneous risk perception, hierarchical assessment, and closed-loop feedback control mechanisms. This significantly improves the safety, stability, and cost-effectiveness of offshore energy access systems. This innovative solution not only enhances the intelligence level of offshore energy access but also provides a viable technical path for future offshore energy system development. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] The present invention is further described with reference to the accompanying drawings. However, the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative effort.

[0086] Figure 1 This is a flow chart of an energy access method based on offshore risk construction according to the present invention. DETAILED DESCRIPTION

[0087] The following describes embodiments of the present invention in detail. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended only to explain the present invention and are not to be construed as limiting the present invention.

[0088] like Figure 1 As shown, an embodiment of the present invention provides an energy access method based on offshore risk construction, the method comprising:

[0089] S1. Collect data sources, build a risk assessment model for each data source to perform risk assessment, and obtain a risk assessment value for each data source; wherein the data sources include meteorological data, ocean data, and equipment data.

[0090] Specifically, offshore energy access systems face data from diverse fields, including meteorological data (such as wind speed and temperature), oceanographic data (such as wave height and tidal cycle), equipment status data (such as generator operating status and load data), and power grid data (such as voltage, current, and load). This data is heterogeneous and time-varying, meaning that data sources, measurement methods, timescales, and accuracy vary. Traditional data fusion methods often fail to effectively process this complex and diverse data, especially given the high real-time and security requirements of offshore energy access, making it difficult to accurately assess and predict potential risks.

[0091] To overcome this problem, the present invention designs a dynamic risk assessment model based on adaptive weighted fusion of multi-source heterogeneous data. The model can integrate risk information from different data sources in real time, dynamically adjust its weights, and output an accurate risk assessment value, thereby providing a reliable basis for subsequent energy access decisions.

[0092] First, the present invention defines a risk assessment model for each data source. Assuming that meteorological data is X1, ocean data is X2, and equipment data is X3, the risk assessment value of each data source is R(X i ) and uses historical volatility to measure its risk contribution. By dynamically weighting the risk assessment values ​​of different data sources, the present invention can assess and predict the potential risks of the system in real time.

[0093] Specifically, meteorological data (such as wind speed, temperature, etc.) has a direct impact on the safety and efficiency of offshore power generation systems. The present invention uses the standard deviation of meteorological data to assess its volatility and defines the risk assessment value R(X1) of meteorological data:

[0094]

[0095] Among them, α1 is a hyperparameter related to the sensitivity of meteorological data fluctuations, and std(X1) is the standard deviation of meteorological data X1, which indicates the fluctuation range of factors such as wind speed and temperature.

[0096] R(X1): Risk assessment value of meteorological data, reflecting the potential risk of meteorological data to the system.

[0097] α1: Sensitivity coefficient of meteorological data to risk assessment, reflecting the response of the model to volatility.

[0098] std(X1): standard deviation of meteorological data, indicating the volatility of meteorological data.

[0099] Specifically, ocean data (such as wave height, tidal changes, etc.) directly affects equipment stability. The present invention uses the standard deviation of ocean data to calculate its risk assessment value R(X2), which is as follows:

[0100]

[0101] Among them, α2 is a hyperparameter related to the sensitivity of ocean environmental fluctuations, and std(X2) is the standard deviation of the ocean data X2, which represents the fluctuation range of environmental factors such as wave height and tide.

[0102] R(X2): Risk assessment value of ocean data, reflecting the potential risk of ocean data to the system.

[0103] α2: Sensitivity coefficient of ocean data to risk assessment, reflecting the response of the model to volatility.

[0104] std(X2): standard deviation of ocean data, indicating the volatility of ocean data.

[0105] Specifically, equipment status data (such as load, health status, etc.) directly affects the reliability of the energy access system. This paper uses the standard deviation of equipment data to evaluate its volatility and defines the equipment risk assessment value R(X3):

[0106]

[0107] Where α3 is a hyperparameter related to the equipment failure history, and std(X3) is the standard deviation of the equipment data X3, which indicates the fluctuation range of equipment load and health status.

[0108] R(X3): Risk assessment value of device data, reflecting the potential risk of device data to the system.

[0109] α3: Sensitivity coefficient of equipment data to risk assessment, reflecting the response of the model to volatility.

[0110] std(X3): standard deviation of the device data, indicating the volatility of the device data.

[0111] S2. Perform weighted fusion on the risk assessment values ​​of each data source to obtain an overall risk assessment value. Based on the overall risk assessment value, determine whether an emergency response is needed: If the overall risk assessment value exceeds the set risk threshold, the system's emergency response mechanism is triggered to ensure stable system operation.

[0112] Specifically, based on the historical accuracy and real-time volatility of each data source, the present invention calculates weighted factors w1, w2, and w3 for each data source, thereby dynamically adjusting the contribution of each data source to the overall risk assessment. Specifically, the weights w1, w2, and w3 can be calculated using the following formula:

[0113]

[0114] Where σ1, σ2, and σ3 are the historical standard deviations of meteorological data, ocean data, and equipment data, respectively, representing the volatility of each data source. Through this weighting strategy, the present invention can ensure that the risk assessment value of each data source contributes dynamically to the overall risk assessment and can adapt to environmental changes in real time. w1, w2, w3: Dynamic weighting coefficients of data sources X1, X2, and X3, representing the weight of the risk assessment of each data source in the overall assessment. σ1, σ2, σ3: Historical standard deviations of meteorological data, ocean data, and equipment data, used to calculate the weight of each data source.

[0115] Furthermore, the risk assessment value R(X i ) and the corresponding weight w i Perform weighted summation to obtain the overall risk assessment value R total , the formula is as follows:

[0116] R total =w1·R(X1)+w2·R(X2)+w3·R(X3) (5)

[0117] Among them, R total : The overall risk assessment value of the system, reflecting the overall risk level of the system under current conditions. w1, w2, w3: The weighting coefficients of meteorological data, ocean data, and equipment data, indicating the contribution of different data sources to the overall risk.

[0118] Furthermore, according to the overall risk assessment value R total , the system will determine whether emergency response is needed. total If the set risk threshold is exceeded, the system's emergency response mechanism (such as equipment switching, load scheduling, etc.) will be triggered to ensure stable operation of the system.

[0119] S3. Build a real-time feedback mechanism to optimize the risk assessment model based on actual operation results.

[0120] Specifically, after the system executes an energy access decision, the present invention collects various data in real time (including equipment operating status, grid load, and environmental data) and feeds this data into the risk assessment model. This feedback allows the present invention to adjust the model's hyperparameters (e.g., α1, α2, and α3) to optimize the risk assessment process.

[0121] Based on real-time feedback data, the present invention retrains the risk assessment model to ensure that it can accurately reflect changes in the current environment. During the retraining process, the present invention adjusts model parameters based on newly acquired data to optimize assessment accuracy and response efficiency.

[0122] Through continuous feedback and optimization, the system can adapt to changes in complex environments, thereby providing more accurate risk assessment and decision support.

[0123] Through the above steps, the present invention has designed a framework that can assess the risks of offshore energy access systems in real time and make timely decisions and emergency responses based on the risk assessment results. This framework can dynamically adapt to environmental changes and optimize the risk assessment and decision-making process to ensure the safe and stable operation of the system.

[0124] S4. Design an adaptive scheduling decision mechanism to dynamically adjust the energy access strategy according to the overall risk assessment value and calculate the final scheduling decision.

[0125] Specifically, an adaptive scheduling decision function D(R total ), whose core purpose is to total The value of is used to adjust the system's energy access or load scheduling strategy.

[0126] When R total Less than the preset risk threshold R threshold When , it means the system is in a low-risk state and energy access can be more flexible;

[0127] When R total Greater than or equal to R threshold When , it means that the system is in a high-risk state and energy access needs to be restricted.

[0128] Specifically, to implement this scheduling decision mechanism, the present invention defines the following scheduling decision function:

[0129] D(R total )=f(R total )=λ1R total +λ2 (6)

[0130] Among them, f(R total ) represents the final output of the scheduling decision, λ1 and λ2 are the adjustment coefficients of the scheduling strategy, which determine the sensitivity of energy access. total Calculated in step 1, it reflects the comprehensive risk of the system. The formula here mainly determines the total When it is high, the output of the scheduling function will quickly tend to a conservative state.

[0131] Among them, D(R total ):Scheduling decision, indicating the change in energy access, which may be a negative value indicating a decrease in access, and a positive value indicating an increase in access. λ1:Scheduling function sensitivity coefficient, which determines the system's response speed to risk. λ2:Scheduling function baseline value, which reflects the system's basic scheduling amount under low-risk conditions. R total : Comprehensive risk assessment value, reflecting the current overall risk level of the system.

[0132] Furthermore, to make the system more flexible in responding to different risk situations, the present invention introduces an adaptive feedback mechanism. This mechanism dynamically adjusts the parameters λ1 and λ2 in the scheduling function based on the actual operation results after each scheduling. The system performs adaptive adjustments through the following steps:

[0133] Feedback input: The system monitors the real-time results of scheduling, such as equipment load, energy utilization efficiency, and equipment health status, and generates feedback signals.

[0134] Parameter update: If the feedback signal indicates that the scheduling effect is not ideal (such as overload, equipment failure or energy shortage), the scheduling parameters λ1 and λ2 will be adjusted to adapt to the current system status.

[0135] The adjustment formula can be expressed as:

[0136] λ1=λ1+Δλ1 and λ2=λ2+Δλ2 (7)

[0137] Among them, Δλ1 and Δλ2 are the adjustment amounts calculated based on the feedback signal. For example, when the system load is too high, Δλ1 can be reduced to reduce the impact on R total λ1: The adjustment coefficient of the dispatch sensitivity, which is dynamically adjusted according to the feedback results. λ2: The adjustment amount of the dispatch baseline value, which corrects the baseline dispatch amount in real time.

[0138] This adaptive feedback mechanism enables the scheduling system to not only respond to initial risk assessments but also optimize based on real-time status.

[0139] Furthermore, in actual offshore energy access systems, in addition to the overall risk of the system (R total ), it is also necessary to consider the influence of external environmental factors, such as weather changes, equipment failures, etc. In order to improve the accuracy of system scheduling, the present invention introduces a weighting factor of the external environment.

[0140] Set the external risk factor as R ext , which can be a risk value related to weather changes (such as wind speed, tidal changes, etc.), or a risk value related to the health status of the equipment. On this basis, the present invention introduces external risk factors into the scheduling decision function, so that the scheduling decision not only depends on the overall risk of the system, but also takes into account the potential impact of the external environment on the system. Therefore, the scheduling decision function is updated as follows:

[0141] D(R total ,R ext )=(λ1R total +λ2)·w ext (8)

[0142] Among them, w ext It is the weighted coefficient of external environmental risk, reflecting the influence of external factors on scheduling decision. ext It can be dynamically adjusted based on real-time monitoring data to increase the system's adaptability to changes in the external environment. ext :External environmental risk factors, which may include weather changes, equipment failures and other external factors. ext : External environment weighting coefficient, which indicates the influence of external factors on scheduling decisions.

[0143] Furthermore, the ultimate goal of scheduling decisions is to ensure that the energy access system remains balanced during operation, avoiding overload, energy waste, or system failure. To this end, the present invention monitors scheduling results in real time and compares them with expected targets. If actual operation deviates from expectations, the system adjusts scheduling decisions in real time.

[0144] The scheduling results (such as energy access, load distribution, equipment health status, etc.) will be transmitted back to the system control center through a real-time feedback mechanism, and the scheduling parameters will be updated based on the real-time data to ensure that the system always operates in the optimal state.

[0145] Through this approach, adaptive scheduling decisions based on risk assessment models can flexibly respond to energy access demands under varying risk conditions, ensuring system security, stability, and efficiency. This approach not only makes real-time decisions based on overall risk but also incorporates external environmental factors for weighted optimization, thereby improving the system's adaptability and robustness.

[0146] S5. By introducing closed-loop feedback control and scheduling strategy optimization, it is ensured that the system can dynamically adjust according to the real-time feedback of scheduling decisions during actual operation to obtain optimized scheduling decisions.

[0147] Specifically, the scheduling decision function D(R total ,R ext ) gives a preliminary energy access strategy, but the system may encounter unexpected changes during operation (such as sudden fluctuations in energy demand, changes in equipment load, etc.), so it is necessary to make adjustments through a closed-loop feedback control mechanism to maintain the adaptability and stability of the scheduling strategy.

[0148] This invention introduces a feedback factor, β, which modifies the initial scheduling decision based on real-time feedback, ensuring that the system can respond promptly to external disturbances. The core idea of ​​the closed-loop control model is to continuously adjust the system's energy access strategy through a feedback mechanism, so that the actual scheduling behavior can maximize the target state. The adjusted scheduling decision function is:

[0149] D opt (R total ,R ext ,β)=D(R total ,R ext )·(1+β) (9)

[0150] Among them, D opt (R total ,R ext ,β) is the adjusted scheduling decision, which is adjusted based on the feedback factor β. total ,R ext ) is the initial scheduling decision obtained in step 2. β is a feedback factor used to adjust the flexibility of the scheduling decision.

[0151] Furthermore, in order to enable scheduling decisions to flexibly respond to real-time changes, the feedback factor β is calculated based on the deviation between the actual operating state of the system and the target. Here, the present invention defines an error function E that represents the gap between the actual scheduling behavior and the target. The error function is calculated as follows:

[0152] E=|D actual -D expected |

[0153] Among them, D actualIt is the scheduling decision actually executed by the system in the current cycle, that is, the actual amount of energy access. expected It is the expected scheduling target of the system (for example, the target value set according to equipment load or energy utilization, etc.).

[0154] In order to adjust β according to the error E, the present invention defines an update rule to adjust the scheduling strategy by considering the impact of the error on the feedback factor:

[0155]

[0156] Where α is the error feedback sensitivity coefficient, which is used to adjust the direct impact of the error E on β. is the rate of change of the error, reflecting its temporal trend. λ is the time response factor, regulating the dynamic response of the error to β. This specialized β update rule not only considers the current error but also the temporal trend of the error, ensuring that the system can quickly adapt to changing trends and adjust scheduling decisions.

[0157] Furthermore, in order to ensure the multi-dimensional optimization of scheduling decisions, the present invention not only focuses on the error E, but also needs to consider other key performance indicators, such as the load L of the equipment, the stability S of the system and the energy efficiency C. Therefore, the present invention designs a multi-objective optimization function Taking all these factors into consideration, the scheduling decision is comprehensively evaluated and optimized. The specific multi-objective optimization function is as follows:

[0158]

[0159] Among them, w1, w2, w3, and w4 are weight coefficients used to balance the priorities of various objectives. E is the error, which indicates the deviation between the actual and expected. L is the load, which indicates the current load of the equipment. S is the system stability, which reflects the stability level of the system under the current scheduling. C is the energy efficiency, which indicates the efficiency of energy utilization. In this way, the present invention not only optimizes the error term in the scheduling decision, but also takes into account multiple factors such as load, stability, and energy efficiency to ensure that the overall performance of the system is optimized.

[0160] Furthermore, to further improve the robustness of the scheduling strategy, the present invention introduces a regularization term R to control the volatility of the scheduling decision. Excessive volatility may lead to system instability, so the present invention adds a regularization term to the optimization objective to limit the range of variation of the scheduling decision:

[0161]

[0162] Among them, R(D opt ) is the change in scheduling decision, which is calculated as:

[0163] R(D opt )=|D opt (t)-D opt (t-1)|

[0164] It represents the difference between the current scheduling decision and the scheduling decision at the previous moment.

[0165] γ is the regularization coefficient, which controls the magnitude of scheduling decision fluctuations. opt ), the present invention can avoid frequent changes in scheduling decisions and reduce the instability caused by excessive adjustments. This regularization term is particularly suitable for systems such as energy access or load scheduling that require stability and gradual transition.

[0166] Furthermore, after each scheduling decision is adjusted, the system implements the new scheduling strategy, executing energy access or load scheduling. During operation, the system monitors various indicators in real time and sends feedback data back to the control system. Based on this real-time feedback, the system further adjusts the feedback factor β, further optimizes the scheduling decision, and executes the next round of scheduling operations.

[0167] Through continuous closed-loop feedback, the system can continuously optimize scheduling strategies in complex and dynamic environments to ensure long-term operational stability and efficiency.

[0168] S6. Comprehensively evaluate the optimized scheduling decision with multiple evaluation indicators, and provide intelligent decision support to decision makers based on this.

[0169] Specifically, to comprehensively evaluate the effectiveness of scheduling decisions, this paper matches key factors such as the system's operating status, load, and stability with multiple preset objectives, ultimately generating a comprehensive evaluation score. This paper proposes an evaluation model based on multi-objective optimization that considers not only the current scheduling decision but also potential future changes and uncertainties, ensuring the accuracy of the evaluation results and the adaptability of the system.

[0170] The present invention defines a comprehensive evaluation function Q(D opt ), used to quantify the performance of each scheduling decision:

[0171]

[0172] Among them, Q(D opt ) is the comprehensive evaluation score, which represents the comprehensive effect of the scheduling decision. i is the objective function f i (D opt ) is the weight coefficient of the total evaluation score. i (D opt ) is the evaluation index function, which represents the contribution of the i-th evaluation index to the scheduling decision.

[0173] Furthermore, in order to ensure a comprehensive evaluation of multiple dimensions, the present invention designs multiple evaluation indicators, covering different key performance areas, including but not limited to system load, energy consumption, operating cost and stability. Each evaluation indicator function f i (D opt ) is designed based on the scheduling decision D obtained in the previous step opt And related system status. Specifically, the evaluation indicators can be the following categories:

[0174] System load L(D opt ): Indicates the load level of the device under the current scheduling decision. If the load is too high, the system may be overloaded; if the load is too low, there may be a waste of resources.

[0175]

[0176] Among them, L j (D opt ) is the load level of the jth device, which represents the actual load of the device under the scheduling decision. m is the number of devices.

[0177] Energy efficiency C(D opt ): represents the energy utilization efficiency under the scheduling decision, taking into account the relationship between energy input and output.

[0178]

[0179] Among them, E out (D opt ) is the actual energy output under the scheduling decision. E in (D opt ) is the amount of energy input.

[0180] Stability S(D opt ) represents the impact of scheduling decisions on system stability. Scheduling decisions that are too high or too low may lead to system instability. This can be measured by monitoring the volatility of system operation.

[0181]

[0182] Among them, D opt (t) is the scheduling decision at time t. T is the length of the evaluation period.

[0183] Furthermore, based on the evaluation model, the present invention needs to be evaluated according to the comprehensive evaluation score Q(D opt ) provides support for decision makers. Based on the scheduling decision score, the present invention proposes a decision recommendation system. The system provides support in the following ways:

[0184] Scenario simulation and prediction: Based on the current comprehensive assessment score Q(D opt ), the system simulates future operating scenarios under different decisions and generates predictions for possible scheduling strategies. These predictions are based on the model’s historical data and analysis of future trends.

[0185] Intelligent recommendation algorithm: The system uses machine learning models (such as regression analysis, decision trees, and reinforcement learning) to intelligently recommend optimal scheduling decisions based on historical data, evaluation scores, and scenario simulation results. For example, given the current system state, it recommends a scheduling strategy that is most likely to achieve optimal system stability and energy efficiency.

[0186] Decision Visualization: Through a visual interface displaying the evaluation scores and recommended strategies for different scheduling decisions, decision makers can intuitively understand the advantages and risks of different options and make the best choice. The system not only provides optimized decisions but also presents decision trends based on historical data and real-time feedback.

[0187] Furthermore, due to the large amount of uncertainty in the actual operating environment (such as equipment failure, energy price fluctuations, etc.), traditional decision support systems may not be able to cope with these dynamic changes. Therefore, the present invention introduces an additional risk assessment module R (D opt ), which is used to consider the impact of these uncertainties on scheduling decisions.

[0188] The present invention defines a risk assessment function R(D opt ), used to quantify the impact of uncertainty on decision quality:

[0189]

[0190] Among them, R(D opt ) is the risk assessment score, which indicates the impact of uncertainty on scheduling decisions. opt ) is an evaluation function that represents the risk level under different scheduling decisions. represents the expected value of the evaluation function, which calculates the weighted average value under all possible risk scenarios.

[0191] This module can help decision makers identify potential risks and provide corresponding warnings and adjustment suggestions when the risks are high.

[0192] Ultimately, the decision support provided by the system goes beyond static recommendations based on the evaluation model. This invention also continuously adjusts the decision support model through a real-time feedback mechanism. During implementation, the system monitors the actual effectiveness of decision execution and optimizes the comprehensive evaluation function and recommendation algorithm based on this feedback, thereby ensuring system adaptability and decision quality.

[0193] Through continuous feedback and optimization, the system forms a closed-loop mechanism, which enables scheduling decisions to be continuously adjusted according to actual execution conditions, ensuring the stability and efficiency of the system in a dynamic environment.

[0194] An embodiment of the present invention further provides an energy access system based on offshore risk construction, the system comprising:

[0195] A data source acquisition unit is used to acquire data sources, construct a risk assessment model for each data source to perform risk assessment, and obtain a risk assessment value for each data source; wherein the data sources include meteorological data, ocean data, and equipment data;

[0196] The risk assessment unit is used to perform weighted fusion on the risk assessment values ​​of each data source to obtain an overall risk assessment value. Based on the overall risk assessment value, it is determined whether an emergency response is required:

[0197] If the overall risk assessment value exceeds the set risk threshold, the system's emergency response mechanism will be triggered to ensure stable system operation;

[0198] Evaluation and optimization unit, used to build a real-time feedback mechanism and optimize the risk assessment model based on actual operation results;

[0199] The energy access analysis unit is used to design an adaptive scheduling decision mechanism, dynamically adjust the energy access strategy according to the overall risk assessment value, and calculate the final scheduling decision, including:

[0200] Design the initial adaptive scheduling decision function to adjust the system's energy access or load scheduling strategy according to the overall risk assessment value:

[0201] When the overall risk assessment value is less than the preset risk threshold, it means that the system is in a low-risk state and energy access is flexible;

[0202] When the overall risk assessment value is greater than or equal to the preset risk threshold, it means that the system is in a high-risk state and energy access needs to be restricted;

[0203] Design an adaptive feedback mechanism to dynamically adjust the adjustment coefficient of the scheduling sensitivity of the initial adaptive scheduling decision function and the adjustment amount of the scheduling benchmark value based on the actual operation effect after each scheduling;

[0204] The initial adaptive scheduling decision function after dynamic adjustment is modified by combining external risk factors to obtain the scheduling decision function, which is expressed as:

[0205] D(R total ,R ext )=(λ1R total +λ2)·w ext

[0206] Among them, D(R total ,R ext ) is the scheduling decision, which indicates the change in energy access. A negative value indicates a decrease in access, while a positive value indicates an increase in access. R total is the overall risk assessment value, R ext is the external environmental risk factor, w ext is the weighted coefficient of external environmental risk, λ1 is the adjustment coefficient of scheduling sensitivity, and λ2 is the adjustment amount of the scheduling benchmark value;

[0207] The access optimization unit is used to introduce closed-loop feedback control and scheduling strategy optimization to ensure that the system can dynamically adjust according to the real-time feedback of scheduling decisions during actual operation and obtain optimized scheduling decisions;

[0208] Personalized support unit, used to comprehensively evaluate the optimized scheduling decision with multiple evaluation indicators, and provide intelligent decision support to decision makers based on this

[0209] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0210] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a division of logical functions. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0211] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0212] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the claims and their equivalents.

Claims

1. An energy access method based on offshore risk construction, characterized in that: The method comprises the following steps: S1. Collect data sources, build a risk assessment model for each data source to perform risk assessment, and obtain a risk assessment value for each data source; wherein the data sources include meteorological data, ocean data, and equipment data; S2. Perform weighted fusion on the risk assessment values ​​of each data source to obtain an overall risk assessment value. Based on the overall risk assessment value, determine whether an emergency response is required: If the overall risk assessment value exceeds the set risk threshold, the system's emergency response mechanism will be triggered to ensure stable system operation; S3. Build a real-time feedback mechanism to optimize the risk assessment model based on actual operation results; S4. Design an adaptive scheduling decision mechanism to dynamically adjust the energy access strategy based on the overall risk assessment value and calculate the final scheduling decision, including: Design the initial adaptive scheduling decision function to adjust the system's energy access or load scheduling strategy according to the overall risk assessment value: When the overall risk assessment value is less than the preset risk threshold, it means that the system is in a low-risk state and energy access is flexible; When the overall risk assessment value is greater than or equal to the preset risk threshold, it means that the system is in a high-risk state and energy access needs to be restricted; Design an adaptive feedback mechanism to dynamically adjust the adjustment coefficient of the scheduling sensitivity of the initial adaptive scheduling decision function and the adjustment amount of the scheduling benchmark value based on the actual operation effect after each scheduling; The initial adaptive scheduling decision function after dynamic adjustment is modified by combining external risk factors to obtain the scheduling decision function, which is expressed as: D(R total ,R ext )=(λ1R total +λ2)·w ext Among them, D(R total ,R ext ) is the scheduling decision, which indicates the change in energy access. A negative value indicates a decrease in access, while a positive value indicates an increase in access. R total is the overall risk assessment value, R ext is the external environmental risk factor, w ext is the weighted coefficient of external environmental risk, λ1 is the adjustment coefficient of scheduling sensitivity, and λ2 is the adjustment amount of the scheduling benchmark value; S5. By introducing closed-loop feedback control and scheduling strategy optimization, we ensure that the system can dynamically adjust according to the real-time feedback of scheduling decisions during actual operation to obtain optimized scheduling decisions; S6. Comprehensively evaluate the optimized scheduling decision with multiple evaluation indicators, and provide intelligent decision support to decision makers based on this.

2. The energy access method based on offshore risk construction according to claim 1, characterized in that: The risk assessment model includes a meteorological data risk assessment model, an ocean data risk assessment model, and an equipment data risk assessment model, which are specifically expressed as follows: Among them, α1 is a hyperparameter related to the sensitivity of meteorological data fluctuations, std(X1) is the standard deviation of meteorological data X1, α2 is a hyperparameter related to the sensitivity of marine environment fluctuations, std(X2) is the standard deviation of marine data X2, α3 is a hyperparameter related to the equipment failure history, std(X3) is the standard deviation of equipment data X3.

3. The energy access method based on offshore risk construction according to claim 2 is characterized in that: The weighted fusion of the risk assessment values ​​of each data source to obtain the overall risk assessment value specifically includes: Based on the historical accuracy and real-time volatility of each data source, weighting factors w1, w2, and w3 are calculated for each data source to dynamically adjust the contribution of each data source to the overall risk assessment value. The calculation is as follows: Among them, σ1, σ2, and σ3 are the historical standard deviations of meteorological data, ocean data, and equipment data, respectively, indicating the volatility of each data source; w1, w2, and w3 are the dynamic weighting coefficients of data sources X1, X2, and X3, respectively, indicating the weight of the risk assessment of each data source in the overall assessment; The risk assessment value R(X i ) and the corresponding weight w i Perform weighted summation to obtain the overall risk assessment value R total , the formula is as follows: R total =w1·R(X1)+w2·R(X2)+w3·R(X3) Among them, R total It is the overall risk assessment value, reflecting the overall risk level of the system under current conditions.

4. The energy access method based on offshore risk construction according to claim 1 is characterized in that: The real-time feedback mechanism is constructed to optimize the risk assessment model based on actual operation results, specifically including: After the system makes an energy access decision, it collects various data in real time and feeds this data into various risk assessment models. By using this feedback data, the hyperparameters in the model are adjusted, thereby optimizing the risk assessment process. Based on real-time feedback data, the risk assessment model is retrained to reflect changes in the current environment.

5. The energy access method based on offshore risk construction according to claim 1 is characterized in that: The adaptive feedback mechanism specifically includes: Monitor the real-time results after scheduling and generate feedback signals. If the feedback signals indicate that the scheduling effect is not ideal, adjust the adjustment coefficient of the scheduling sensitivity and the adjustment amount of the scheduling reference value, which can be expressed as: λ1=λ1+Δλ1andλ2=λ2+Δλ2 Among them, Δλ1 and Δλ2 are adjustment amounts calculated based on the feedback signal. λ1 is the adjustment coefficient of the scheduling sensitivity, which is dynamically adjusted according to the feedback results; λ2 is the adjustment amount of the scheduling benchmark value, which corrects the benchmark scheduling amount in real time.

6. The energy access method based on offshore risk construction according to claim 1 is characterized in that: Said S5 specifically includes: The feedback factor is introduced to modify the initial scheduling decision based on real-time feedback to ensure that the system can respond to external disturbances in a timely manner. The adjusted scheduling decision function is: D opt (R total ,R ext ,β)=D(R total ,R ext )·(1+b) Among them, D opt (R total ,R ext ,β) is the adjusted scheduling decision, which adjusts the initial decision according to the feedback factor β; Define the error function E, which represents the gap between the actual scheduling behavior and the target; Define an update rule to adjust β according to the error function E, and adjust the scheduling strategy according to the impact of the error on the feedback factor; wherein the update rule is expressed as: Where α is the error feedback sensitivity coefficient, which is used to adjust the direct impact of the error E on β; is the rate of change of the error, reflecting the trend of the error over time; λ is the time response coefficient, which is used to adjust the dynamic response degree of the error change to β; Combining the error E and the feedback factor β, a multi-objective optimization function is designed to comprehensively evaluate and optimize the scheduling decision. opt ), expressed as: O(D opt )=w1·E+w2·L+w3·S+w4·C Among them, w1, w2, w3, and w4 are weight coefficients used to balance the priorities of various objectives; E is the error, which indicates the deviation between the actual and expected; L is the load, which indicates the current load of the equipment; S is the system stability, which reflects the stability level of the system under the current scheduling; C is the energy efficiency, which indicates the efficiency of energy utilization.

7. The energy access method based on offshore risk construction according to claim 6 is characterized in that: In the multi-objective optimization function O(D opt ) adds a regularization term R to control the volatility of scheduling decisions and limit the range of changes in scheduling decisions, which can be expressed as: O regularized (D opt )=O(D opt )+γ·R(D opt ) Among them, O regularized (D opt ) is the regularized multi-objective optimization function, R(D opt ) is the change in scheduling decision, which is calculated as: R(D opt )=|D opt (t)-D opt (t-1)|, which represents the difference between the current scheduling decision and the scheduling decision at the previous moment; γ is the regularization coefficient, which controls the magnitude of the fluctuation of the scheduling decision.

8. The energy access method based on offshore risk construction according to claim 6 is characterized in that: Said S6 specifically includes: An evaluation model based on multi-objective optimization is constructed to obtain a comprehensive evaluation score; wherein the evaluation model based on multi-objective optimization is calculated as follows: Among them, Q(D opt ) is the comprehensive evaluation score, which represents the comprehensive effect of the scheduling decision; w i is the objective function f i (D opt ) is the weight coefficient of the total evaluation score; f i (D opt ) is the evaluation index function, which represents the contribution of the i-th evaluation index to the scheduling decision; Multiple evaluation metrics were designed, covering different key performance areas, including but not limited to system load, energy consumption, operating cost, and stability; Provide support to decision makers based on comprehensive assessment scores, including: Based on the current comprehensive evaluation score, simulate future operating scenarios under different decisions and generate predictions of possible scheduling strategies; Based on historical data, comprehensive evaluation scores, and scenario simulation results, machine learning models are used to intelligently recommend optimal scheduling decisions. The evaluation scores and recommended strategies of different scheduling decisions are displayed through a visual interface.

9. The energy access method based on offshore risk construction according to claim 8, characterized in that: According to the uncertainty in the actual operating environment, the risk assessment function R(D opt ), which is used to consider the impact of these uncertainties on scheduling decisions, is expressed as: R(D opt )=E[f(D opt )] Among them, R(D opt ) is the risk assessment score, which indicates the impact of uncertainty on scheduling decisions; f(D opt ) is the evaluation function, which represents the risk level under different scheduling decisions; E[f(D opt )] represents the expected value of the evaluation function, and the weighted average value under all possible risk scenarios is calculated.

10. An energy access system based on offshore risk construction, characterized in that: The system comprises: A data source acquisition unit is used to acquire data sources, construct a risk assessment model for each data source to perform risk assessment, and obtain a risk assessment value for each data source; wherein the data sources include meteorological data, ocean data, and equipment data; The risk assessment unit is used to perform weighted fusion on the risk assessment values ​​of each data source to obtain an overall risk assessment value. Based on the overall risk assessment value, it is determined whether an emergency response is required: If the overall risk assessment value exceeds the set risk threshold, the system's emergency response mechanism will be triggered to ensure stable system operation; The assessment and optimization unit is used to build a real-time feedback mechanism and optimize the risk assessment model based on actual operation results, including: an energy access analysis unit, configured to design an adaptive scheduling decision mechanism, dynamically adjust the energy access strategy according to the overall risk assessment value, and calculate a final scheduling decision; Design the initial adaptive scheduling decision function to adjust the system's energy access or load scheduling strategy according to the overall risk assessment value: When the overall risk assessment value is less than the preset risk threshold, it means that the system is in a low-risk state and energy access is flexible; When the overall risk assessment value is greater than or equal to the preset risk threshold, it means that the system is in a high-risk state and energy access needs to be restricted; Design an adaptive feedback mechanism to dynamically adjust the adjustment coefficient of the scheduling sensitivity of the initial adaptive scheduling decision function and the adjustment amount of the scheduling benchmark value based on the actual operation effect after each scheduling; The initial adaptive scheduling decision function after dynamic adjustment is modified by combining external risk factors to obtain the scheduling decision function, which is expressed as: D(R total ,R ext )=(λ1R total +λ2)·w ext Among them, D(R total ,R ext ) is the scheduling decision, which indicates the change in energy access. A negative value indicates a decrease in access, while a positive value indicates an increase in access. R total is the overall risk assessment value, R ext is the external environmental risk factor, w ext is the weighted coefficient of external environmental risk, λ1 is the adjustment coefficient of scheduling sensitivity, and λ2 is the adjustment amount of the scheduling benchmark value; The access optimization unit is used to introduce closed-loop feedback control and scheduling strategy optimization to ensure that the system can dynamically adjust according to the real-time feedback of scheduling decisions during actual operation and obtain optimized scheduling decisions; The personalized support unit is used to comprehensively evaluate the optimized scheduling decisions with multiple evaluation indicators, and based on this, provide intelligent decision support to decision makers.

Images