Multi-energy coordination optimization method and device based on port power load characteristics

Abstract

The present invention relates to the technical field of multi-energy complementarity and optimal configuration, and discloses a multi-energy coordinated optimization method and device based on the characteristics of port electricity loads. Specifically, it includes: obtaining high-resolution power parameters of the load and power parameters supplied by renewable resources; matching the high-resolution power parameters of the load and the power parameters supplied by renewable resources, determining the matching matrix of the power required by the load and the power supplied by renewable resources, and determining the power supply resource allocation strategy of the multi-energy storage system based on the matching matrix; based on the objective function of the power supply resource allocation strategy, coordinating and optimizing the power supply resource allocation strategy to determine a dynamic optimization strategy. By coordinating and optimizing the power supply resource allocation strategy based on the objective function considered in the present invention, the power supply cost and the waste of power supply resources can be reduced, thereby enhancing the stability and reliability of the port energy system.

Description

Technical Field

[0001] The present disclosure relates to the technical field of multi-energy complementarity and optimal configuration, and in particular to a multi-energy coordination optimization method and device based on port electricity load characteristics. Background Art

[0002] With the increasing global demand for clean energy and the diversification of power supply systems, optimizing hybrid energy output based on port load characteristics has become a key research focus.

[0003] As key nodes in the global supply chain, ports have unique energy demand characteristics, which are manifested in round-the-clock uninterrupted operations, sudden load fluctuations caused by container cranes, and strict reliability requirements of cold chain facilities.

[0004] Related technologies have proposed supplementary power supply solutions and optimization models for integrated clean energy systems tailored to the specific load characteristics of ports. However, these solutions and optimization models suffer from fundamental flaws, such as high power supply costs and unstable output. Summary of the Invention

[0005] In order to solve the above technical problems, the present disclosure provides a multi-energy coordination optimization method based on the characteristics of port power load, including the following steps:

[0006] Obtain high-resolution power parameters of the load and power parameters supplied by renewable resources; match the high-resolution power parameters of the load and the power parameters supplied by renewable resources to determine a matching matrix of the power required by the load and the power supplied by renewable resources, and determine a power supply resource allocation strategy for the multi-energy storage system based on the matching matrix; based on the objective function of the power supply resource allocation strategy, coordinate and optimize the power supply resource allocation strategy to determine a dynamic optimization strategy.

[0007] Furthermore, determining a matching matrix between the power required by the load and the power supplied by renewable resources includes:

[0008] Based on the high-resolution power parameters of the load, load characteristics are extracted, and based on the power parameters supplied by the renewable resources, renewable energy operation characteristics are extracted; a first interaction relationship between the load characteristics and the renewable energy operation is determined, and based on the first interaction relationship, a coupling model of the load and the renewable energy is determined; the coupling model is used to match the load characteristics and the renewable energy operation characteristics to determine a matching matrix of the power required by the load and the power supplied by the renewable resources.

[0009] Furthermore, determining the power supply resource allocation strategy of the multi-energy storage system according to the matching matrix includes:

[0010] Determine the load characteristic type based on the high-resolution power parameters of the load; determine the characteristic quantitative index of the load characteristic type and the renewable resource supply power parameter; determine tidal energy based on solar energy and wind energy from historical meteorological data; use the tidal energy as supplementary power supply during low electricity consumption periods, match the supplementary power supply and the characteristic quantitative index through the matching matrix, and determine the power supply resource allocation strategy of the multi-energy storage system.

[0011] Furthermore, determining the load characteristic type based on the load high-resolution power parameter includes:

[0012] Analyze the high-resolution power parameters of the load to determine the second interaction relationship between the load's operating conditions, equipment energy consumption, and environmental factors and the equipment; and determine the load characteristic type based on the second interaction relationship.

[0013] Furthermore, the determining of the load characteristic type and the characteristic quantitative index of the renewable resource supply power parameter includes:

[0014] Analyze the load characteristic type to determine the load pattern quantification index, the load time regularity quantification index and the load power consumption fluctuation quantification index; determine a second interaction relationship between the load pattern quantification index, the load time regularity quantification index, the load power consumption fluctuation quantification index and the renewable resource supply power parameter; analyze and quantify the second interaction relationship through spectral coherence to obtain the characteristic quantification index of the renewable resource.

[0015] Furthermore, the tidal energy is used as a supplementary power supply during the off-peak period, and the supplementary power supply and the characteristic quantitative index are matched through the matching matrix to determine the power supply resource allocation strategy of the multi-energy storage system, including:

[0016] Obtain determined load pattern quantification indicators, load time regularity quantification indicators, and load power consumption fluctuation quantification indicators; based on the load pattern quantification indicators, the load time regularity quantification indicators, and the load power consumption fluctuation quantification indicators, determine the lithium-ion battery configuration parameters, capacitor model parameters, and time parameters for using tidal energy as a supplementary power supply during low-power periods of the multi-energy energy storage system; determine the power supply resource allocation strategy of the multi-energy energy storage system based on the lithium-ion battery configuration parameters, the capacitor model parameters, and the time parameters.

[0017] Furthermore, the objective function of the power supply resource allocation strategy is used to coordinately optimize the power supply resource allocation strategy and determine a dynamic optimization strategy, including:

[0018] The Pareto front resolution is dynamically adjusted based on the adaptive reference point and port size of the third-generation non-dominated sorting genetic algorithm, and the time weight is determined based on the time discount effect of the third-generation non-dominated sorting genetic algorithm; the conflict mechanism between the objectives of the objective function is determined; according to the adjusted Pareto front resolution and the time weight, the conflict mechanism between the objectives is weighed, the objective function is dynamically optimized, and a dynamic optimization strategy for the power supply resource allocation strategy is determined.

[0019] Furthermore, the objective function includes one or more of a levelized cost of electricity objective function, a carbon dioxide emission objective function, and a power shortage objective function.

[0020] Furthermore, the objective function also includes constraints;

[0021] The constraint conditions include one or more of power balance constraints, generator operation limitation constraints, energy storage dynamic characteristic constraints, renewable energy penetration constraints, and equipment and energy coupling constraints.

[0022] The multi-energy coordinated optimization device based on port power load characteristics adopts the multi-energy coordinated optimization method based on port power load characteristics as described in any of the above items, and the device includes:

[0023] An acquisition module is used to obtain high-resolution power parameters of the load and power parameters supplied by renewable resources; a determination module is used to match the high-resolution power parameters of the load and the power parameters supplied by renewable resources, determine the matching matrix of the power required by the load and the power supplied by renewable resources, and determine the power supply resource allocation strategy of the multi-energy storage system based on the matching matrix; an optimization module is used to coordinate and optimize the power supply resource allocation strategy based on the objective function of the power supply resource allocation strategy and determine the dynamic optimization strategy.

[0024] The embodiments of the present disclosure have the following technical effects:

[0025] The multi-energy coordination optimization method based on the port power load characteristics provided in this application obtains high-resolution load power parameters and renewable resource supply power parameters, thereby mechanistically linking the dynamic characteristics of port equipment with the operating characteristics of renewable energy through a matching matrix of the determined load power requirements and renewable resource supply power parameters. This method does not rely on a specific data set, and better matches the operating characteristics of renewable energy with the needs of port equipment. This can increase the proportion of renewable energy in the port energy supply, reduce dependence on traditional fossil energy, and promote the development of ports in a green and low-carbon direction, meeting the requirements of sustainable development. The power supply resource allocation strategy is coordinated and optimized based on its objective function, and multiple objectives can be considered. The resulting dynamic optimization strategy has low cost and stable power supply, and reduces carbon dioxide emissions, thereby enhancing the stability and reliability of the port energy system. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0027] Figure 1 This is a flow chart of a multi-energy coordination optimization method based on port power load characteristics provided by an embodiment of the present disclosure;

[0028] Figure 2 is a flow chart of a method for determining a power supply resource allocation strategy provided by an embodiment of the present disclosure;

[0029] Figure 3 is a three-dimensional surface graph of annual power demand provided by an embodiment of the present disclosure;

[0030] Figure 4 This is a load energy consumption ratio ring chart generated by IoT sensor data provided by an embodiment of the present disclosure;

[0031] Figure 5 This is an analysis diagram using a clustering algorithm provided by an embodiment of the present disclosure;

[0032] Figure 6 This is a schematic diagram of analyzing load power fluctuation quantitative indicators using kernel density estimation provided by an embodiment of the present disclosure;

[0033] Figure 7 This is a schematic diagram of a multi-energy coordination and optimization device based on port electricity load characteristics provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0034] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be described clearly and completely below. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort are considered to be within the scope of this disclosure.

[0035] Global maritime transport carries more than 80% of international trade, necessitating an urgent transformation of port energy systems.

[0036] As key nodes in the global supply chain, ports have unique energy demand characteristics, characterized by 24 / 7 operations, sudden load fluctuations caused by container cranes, and stringent reliability requirements for cold chain facilities. Automation-driven energy demand in ports has increased by 17% in recent years, yet emissions per TEU-km (TEU-km) have remained stagnant due to their reliance on traditional diesel generators.

[0037] Current technologies for developing port energy systems primarily focus on integrating renewable energy with traditional generators, but this has limitations and could reduce renewable energy generation. Other proposals for integrating offshore wind power with hydrogen storage systems can lead to unstable turbine output. Taking all this into account, these technologies suffer from the following three systemic blind spots.

[0038] (1) Load characteristic analysis often simplifies complex industrial processes into static curves;

[0039] (2) Resource modeling often ignores local meteorological interactions;

[0040] (3) There is a persistent time disconnect between the planning and operation stages.

[0041] Based on the above technical problems, the present disclosure proposes a multi-energy coordinated optimization method and device based on the characteristics of port electricity loads. Through a physical information-driven device-energy coupling model, the supply and demand mismatch matrix of the load high-resolution operation data and the renewable resource data is determined, and the dynamic characteristics of the port equipment are linked to the operating characteristics of renewable energy from a mechanistic point of view. It does not rely on a specific data set, and has better adaptability and generalization capabilities for port equipment and energy systems under different working conditions and different environmental conditions, and can be more reliably applied to actual port operation scenarios. The power supply resource allocation strategy is optimized based on a multi-objective optimization model to determine the dynamic optimization strategy for the coordinated output power supply of the multi-energy storage system. Multiple important objectives are taken into consideration, so that the resulting dynamic optimization strategy for the coordinated output power supply of the multi-energy storage system has low cost and stable power supply, and reduces carbon dioxide emissions, thereby enhancing the stability and reliability of the port energy system.

[0042] Figure 1 This is a flow chart of the multi-energy coordination optimization method based on port power load characteristics provided by the embodiment of the present disclosure. Figure 1 , specifically including the following steps:

[0043] In step S11 , high-resolution load power parameters and renewable resource supply power parameters are acquired.

[0044] In the disclosed embodiment, the acquired high-resolution load power parameters can be high-resolution power parameters for the target automated terminal over the past two years. These parameters include crane motor current (e.g., 3,000 samples per second), refrigerated container thermal cycles acquired through infrared imaging, and ship-to-shore power curves. Renewable resource data can include power supply parameters converted from renewable resources such as solar and wind energy.

[0045] In the present disclosure, it is also possible to apply a filter to remove data values ​​of abnormal load high-resolution power parameters and renewable resource supply power parameters, and synchronize time stamps.

[0046] In step S12, the high-resolution power parameters of the load and the power parameters of the renewable resources are matched to determine the matching matrix of the power required by the load and the power supplied by the renewable resources. Based on the matching matrix, the power supply resource allocation strategy of the multi-energy storage system is determined.

[0047] In the disclosed embodiment, the load curve of the load can be determined based on the high-resolution power parameters of the load, thereby matching the load curve with the curve corresponding to the power parameters supplied by renewable resources to obtain a matching matrix of the power required by the load and the power supplied by renewable resources.

[0048] The matching matrix can also be understood as a difference matching matrix between the power required by the load and the power supplied by the renewable resources, or the matching matrix can be called a mismatch matrix.

[0049] Furthermore, the power supply resource allocation strategy of the multi-energy storage system can be formulated by analyzing the relationship between the matching matrix and the characterized load curve and the irradiance of renewable energy.

[0050] In step S13, based on the objective function of the power supply resource allocation strategy, the power supply resource allocation strategy is coordinated and optimized to determine a dynamic optimization strategy.

[0051] In the disclosed embodiment, the objective function is pre-constructed and can adjust 12 decision variables. For example, it can range from the photovoltaic tilt angle to the ramp rate of the LNG generator, while balancing objectives such as minimizing energy costs and CO2 intensity. An enhanced third-generation non-dominated sorting genetic algorithm is used to determine the dynamic optimization strategy for coordinating the output power supply of the multi-energy storage system. Furthermore, 20% of the multi-objective optimization model population is initialized based on configurations derived from data analysis tools.

[0052] The disclosed embodiment provides a multi-energy coordinated optimization method based on the characteristics of the port's electricity load, which obtains high-resolution load power parameters and renewable resource supply power parameters, thereby mechanistically linking the dynamic characteristics of port equipment with the operating characteristics of renewable energy through a matching matrix of the determined load required power and renewable resource supply power parameters. This method does not rely on a specific data set, and better matches the operating characteristics of renewable energy with the needs of port equipment. This can increase the proportion of renewable energy in the port's energy supply, reduce dependence on traditional fossil energy, and promote the development of ports in a green and low-carbon direction, meeting the requirements of sustainable development. The power supply resource allocation strategy is optimized based on its objective function, and multiple objectives can be considered. The resulting dynamic optimization strategy for the coordinated output power supply of the multi-energy storage system has low cost and stable power supply, and reduces carbon dioxide emissions, thereby enhancing the stability and reliability of the port energy system.

[0053] Furthermore, in the disclosed embodiments, load characteristics can be extracted based on high-resolution load power parameters, and renewable energy operation characteristics can be extracted based on renewable energy supply power parameters. This allows for determining a first interaction relationship between the load characteristics and renewable energy operation, and based on the first interaction relationship, determining a coupling model for the load and renewable energy. The coupling model is then used to match the load characteristics and renewable energy operation characteristics to determine a matching matrix between the load power required and the renewable energy supply power. The load characteristics can be a characterized load curve.

[0054] In the disclosed embodiment, the port load characteristics can be analyzed to determine the two major dimensions of load data classification and feature quantification. By integrating high-precision measured data with advanced statistical methods, the complex coupling relationship between equipment operation dynamics, environmental factors and energy demand patterns can be systematically analyzed to determine the power supply resource allocation strategy. Figure 2 This is a flow chart of the method for determining the power supply resource allocation strategy provided by the embodiment of the present disclosure. Figure 2 , and its implementation methods include the following examples.

[0055] In step S21 , the load characteristic type is determined based on the load high-resolution power parameter.

[0056] In the embodiment of the present disclosure, high-resolution power parameters of the load can be analyzed to determine the second interaction relationship between the load's operating conditions, equipment energy consumption, and environmental factors and the equipment, thereby determining the load feature type based on the second interaction relationship.

[0057] In the present disclosure, the load of the port has typical multi-scale heterogeneity characteristics, so different load high-resolution power parameters have different load characteristics, corresponding to different feature types. Figure 3 It is a three-dimensional surface diagram of annual power demand provided by the embodiment of the present disclosure. Figure 3 Based on 730 days of power parameters from 2019 to 2021 with a time resolution of 15 minutes, the annual power demand three-dimensional surface (month × date × power) reconstructed by cubic spline interpolation clearly shows the following load characteristics:

[0058] (1) The peak and valley characteristics of the load during the day and night are manifested in that the peak power demand (6.8-8.2MW) is strictly synchronized with the ship berthing schedule.

[0059] (2) The seasonal difference in load is manifested in that the base load in winter is 23% higher than that in summer due to the demand for cold chain.

[0060] (3) The characteristics of the impact of extreme load events are manifested as a 35-40% load drop during the typhoon period from July to September.

[0061] Window length Singular Spectrum Analysis (SSA) analyzes the seasonal characteristics of the load and identifies the dominant trend component that influences the seasonal characteristics of the load, which is influenced by the long-term influence of port infrastructure. It also confirms that the seasonal residuals of the seasonal characteristics of the load are strongly correlated with the monsoon humidity cycle.

[0062] By using the isolation forest algorithm to detect anomalies in the seasonal difference characteristics of the load, it was determined that the load trough period (July to September) caused by typhoons showed a load reduction of 35% to 40%.

[0063] In the present disclosure, the load may be electrical equipment such as bridge cranes, cold chain facilities, and ships. Figure 4 This is a circular chart of load energy consumption ratio generated by IoT sensor data provided by the embodiment of the present disclosure. Figure 4 The circular chart, generated based on 18 months of IoT sensor data with a sampling frequency of 3kHz, analyzes the energy consumption contribution of each load, including:

[0064] The energy consumption of bridge cranes accounts for 38±5% of the total energy consumption. The power curve is trapezoidal. When using container lifting, the motor current rises from 0 to 3.8MW within 12 seconds, and the power plateau lasts for 45±8 seconds.

[0065] The energy consumption of cold chain facilities accounts for 29±3%, and the thermodynamic characteristics of the cold chain facilities conform to the constant temperature hysteresis phenomenon described by the lumped capacitance equation.

[0066] The demand pattern of shore power in ports (accounting for 19% of the total load) is strongly correlated with the tonnage of ships and presents a bimodal distribution.

[0067] It should be noted that environmental factors can cause nonlinear disturbances in load energy consumption. The impact mechanism is that when the crosswind speed exceeds 10 m / s, the operating speed of the bridge crane will decrease. At the same time, a tidal surge exceeding 1.5 m will interfere with the shore power connection, causing a 2 to 4 minute power outage in the multi-energy storage system, which will trigger a spike in load energy consumption when power is restored.

[0068] In summary, different load high-resolution power parameters correspond to different load characteristic types, and related steps are further performed according to the load characteristic type.

[0069] In step S22, the load characteristic type and the characteristic quantitative index of the renewable resource supplied power parameter are determined.

[0070] In the embodiment of the present disclosure, it is necessary to convert the original data into executable design parameters, that is, quantitative indicators.

[0071] Furthermore, advanced statistical decomposition methods can be used to analyze load characteristics and determine quantitative indicators for load patterns, load temporal regularity, and load power fluctuation. This allows the identification of a third interaction relationship between these indicators and the power parameters of renewable energy supply. Spectral coherence is used to analyze and quantify this third interaction relationship, yielding a quantitative indicator of renewable energy characteristics.

[0072] Among them, the third interaction relationship is a dynamic correlation. Taking the bridge crane as an example, the operation of the bridge crane and the gust of wind show a coherence of 0.85 in the frequency band of 0.1-0.3Hz.

[0073] Furthermore, the K-means++ clustering algorithm is used to analyze the standardized daily load curve, and the following load characteristic types can be obtained. Figure 5 This is an analysis diagram using a clustering algorithm provided by an embodiment of the present disclosure. Figure 5 ,Cluster 1 is the load bimodal feature under the conventional operation mode, ,including the load peak feature type at 10:00 (7.1MW) and the load peak feature type ,at 15:00 (6.8MW), and the load bimodal feature is synchronized with the ,arrival time of the Panamax ship.

[0074] The load characteristic type of cluster 2 is a load characteristic type in which the midnight load may suddenly increase to a certain value in the emergency operation mode (for example, the demand for refrigerated medicine transportation), and this load characteristic type is a transient load characteristic type that cannot be captured by the monthly average load analysis.

[0075] The load characteristic type of cluster 3 is a load characteristic type that shows periodic fluctuations in the load under the meteorological disturbance mode. For example, the 2-hour periodic fluctuation is caused by the coordinated deceleration of the bridge crane during heavy rain.

[0076] Among them, the quantitative indicators of load time regularity determined include the use of daily imbalance coefficient , with a value range of 0.38 (high volatility day) to 0.62 (stable operation day), and is significantly negatively correlated with the demand for energy storage cycle depth.

[0077] Figure 6 This is a schematic diagram of analyzing the quantitative index of load power fluctuation using kernel density estimation provided by the embodiment of the present disclosure. Figure 6 , using kernel density estimation to quantify load power fluctuation indicators The results show that its peak-peaked and fat-tailed distribution characteristics reveal two typical fluctuation patterns: high-frequency peak fluctuations and medium-period fluctuations.

[0078] In step S23, tidal energy is determined based on solar energy and wind energy from historical meteorological data.

[0079] In step S24, tidal energy is used as a supplementary power supply during the off-peak period of electricity consumption, and the supplementary power supply and characteristic quantitative indicators are matched through a matching matrix to determine the power supply resource allocation strategy.

[0080] In the embodiment of the present disclosure, after determining tidal energy using historical meteorological data, tidal energy can be modeled as a supplementary energy source during the period of low electricity consumption, so that the supplementary energy can be converted into energy and the supplementary power supply can be determined. Further, the determined load pattern quantitative index, load time regularity quantitative index, and load power consumption fluctuation quantitative index are obtained; based on the load pattern quantitative index, load time regularity quantitative index, and load power consumption fluctuation quantitative index, the lithium-ion battery configuration parameters, capacitor model parameters, and time parameters of the multi-energy energy storage system using tidal energy as a supplementary power supply during the period of low electricity consumption are determined. Thus, the power supply resource allocation strategy of the multi-energy energy storage system is determined based on the lithium-ion battery configuration parameters, capacitor model parameters, and time parameters. Among them, the power supply resource allocation strategy includes a mixed power supply of renewable energy and power generation equipment. It can not only reduce energy waste and improve power supply reliability, but also has a significant effect in achieving sustainable port energy management.

[0081] Furthermore, for lithium-ion battery configuration, a derating factor of 1.8 times the manufacturer's cycle life is required, which can be based on the rain flow counting method, the damage degree =The rated power of unit i is suitable for operating conditions with frequent shallow charge and discharge cycles. The supercapacitor is selected to have a capacity design that covers the 95th percentile of load peaks (200kW / ms rate of change). Tidal energy is used to supplement power during off-peak periods, during which there is a 2.3-hour time lag between photovoltaic power generation and the crane load. Tidal energy supplementation (nighttime utilization rate of 18%) is introduced, and a demand-side management strategy is implemented to shift non-critical loads to the photovoltaic power generation period.

[0082] In the disclosed embodiments, the Pareto front resolution can be dynamically adjusted based on the adaptive reference point and port size of the third-generation non-dominated sorting genetic algorithm, and the time weight can be determined based on the time discounting effect of the third-generation non-dominated sorting genetic algorithm. This further determines the inter-objective conflict mechanism of the objective function. Based on the adjusted Pareto front resolution and time weight, the inter-objective conflict mechanism is weighed, the objective function is dynamically optimized, and a dynamic optimization strategy for the coordinated output power supply of the multi-energy storage system is determined.

[0083] In the embodiment of the present disclosure, multiple objective functions and their corresponding constraints can be determined based on balancing economic efficiency, environmental sustainability, and operational reliability, wherein the constraints include physical constraints and / or operational constraints.

[0084] Among them, the levelized cost of electricity objective function is determined based on balanced economics, the carbon dioxide emissions objective function is determined based on environmental sustainability, and the power shortage objective function is determined based on operational reliability.

[0085] The expression of the levelized cost of electricity objective function is as follows:

[0086]

[0087] Where, represents the minimum levelized cost of electricity, Indicates the y The annual investment cost includes the costs of LNG units, energy storage systems, photovoltaics, etc. Indicates the y Annual operating costs, including fuel, maintenance, grid interaction and other operating costs. Indicates the y Annual maintenance costs. Indicates the y The total power supply per year. r represents the discount rate (usually 5-8%). N Indicates the system life cycle (e.g. 20 years).

[0088] The numerator of the formula aggregates lifecycle costs, while the denominator discounts the energy supplied, penalizing systems with high energy losses due to high intermittency. This formula inherently favors deferring investments in energy storage technologies, consistent with projected lithium-ion battery price trends, while penalizing excessive curtailment by reducing the amount of power supplied.

[0089] The expression of the carbon dioxide emissions objective function is as follows:

[0090]

[0091] Where, The minimum value of carbon dioxide emissions, indicating carbon emissions on the power generation side represents the carbon intensity of the i-th generator set (kg-CO2 / kWh), P i,t Indicates the unit Output power (kW) at time period t, implicit carbon emissions from energy storage Energy storage system j Implied carbon emissions per unit cycle (kg-CO2 / kWh), Indicates energy storage j The change in state of charge in time period t is: T Indicates the time period, G Indicates the total number of generator sets, S Represents the total number of energy storage systems.

[0092] This formula covers the carbon emission accounting of the entire life cycle, including direct emissions (generator operation) and indirect emissions (carbon costs implied by charging and discharging of energy storage equipment), avoiding carbon leakage problems, refining the time scale, quantifying dynamic carbon emissions with hourly time steps (t), and accurately matching port load fluctuations. Compatible with Scope3 standard The parameters include upstream emissions from energy storage equipment manufacturing and transportation, which is in line with the international carbon accounting system.

[0093] For example, taking the constraint linkage as an example, when the photovoltaic output is insufficient, the model prioritizes the scheduling of carbon emission intensity. Lower-cost LNG units (such as high-efficiency gas turbines) rather than direct electricity purchase.

[0094] The expression of the power shortage objective function is as follows:

[0095]

[0096] Where, Indicates the minimum value of power shortage. R t Represents the amount of electricity generated by renewable energy, Δt = a time interval of 10 to 15 minutes.L t represents the total load demand (kW) during time period t, Energy storage system j In the period t Discharge power (kW), T Indicates the time period, G Indicates the total number of generator sets, S Represents the total number of energy storage systems.

[0097] This formula includes a risk-averse design, which only penalizes insufficient power supply ( L t > total power supply), ignoring excess generation. It also ensures critical loads by directly quantifying the expected energy shortage (EENS), strictly aligned with port operation continuity requirements. It also includes time granularity optimization, with 15-minute calculations matching port operation cycles (such as ship berthing windows).

[0098] For example, the constraint linkage mechanism is used as an example to illustrate that when a typhoon causes R t During a sudden drop, the model prioritizes discharging through energy storage and quick start units to compensate for the shortfall.

[0099] The mechanisms for conflicting objectives include economic Environmental protection Contradictions and reliability and economical contradiction.

[0100] Economical Environmental protection The paradox can be understood as follows: minimizing LCOE favors low-cost fossil fuels (such as LNG), but this drives up carbon emissions. Mandating an increase in the share of renewable energy would increase system capital expenditure (CAPEX) by 15-25%.

[0101] reliability and economical The contradiction can be understood as follows: reducing EENS requires deploying excess energy storage (e.g., a 30% increase in lithium battery capacity), which results in an 8-12% increase in LCOE. However, redundant energy storage can reduce the high penalties for power outages (typically 100,000-500,000 yuan per outage).

[0102] In the disclosed embodiment, the adaptive reference points of the third-generation non-dominated sorting genetic algorithm can dynamically adjust the Pareto front resolution based on port size. For example, small ports (<100MW) can have 50 reference points, while large ports (≥100MW) can have 100 reference points. This improves computational efficiency by 40% while maintaining solution diversity.

[0103] The time discount effect based on the third-generation non-dominated sorting genetic algorithm introduces time-dependent weights, making recent reliability investments more advantageous than long-term storage, which can lead to a rapid decline in battery costs.

[0104] In this disclosure, the dynamic optimization strategy for coordinating the output power supply of a multi-energy storage system can be as follows:

[0105] In the medium term (2025-2030), a hybrid solution of LNG + energy storage can be adopted (carbon emissions / cost balance point), and in the long term (2030+), photovoltaic + tidal energy + short-term energy storage can be adopted (relying on the reduction of technology costs).

[0106] In this disclosure, physical and / or operational constraints ensure the feasibility of the solution, including: equipment physical limitations, operational safety margins, and special port operation requirements. Specifically, equipment physical limitations may include an LNG unit ramp rate of ≤15MW / min, operational safety margins may include an energy storage SOC operating window of 20-95%, and special port operation requirements may include ensuring continuous shore power supply. Based on these requirements, constraints are determined to include one or more of power balance constraints, generator operational limitations, energy storage dynamic characteristics constraints, renewable energy penetration constraints, and equipment-energy coupling constraints.

[0107] The expression of power balance constraint is as follows:

[0108]

[0109] Where, represents the charging power of energy storage system j in time period t (kW), This equation means that for all time periods t All established, Represents the amount of renewable energy reduction, which is limited to not exceed the time t Renewable energy generation R t This constraint not only ensures that supply and demand are balanced but also incorporates strategic energy reduction measures to prevent grid instability during load spikes caused by cranes.

[0110] The expression of the generator operation limit constraint is as follows:

[0111]

[0112] Where, Indicates the unit Minimum output power (kW), Indicates the unit Maximum output power (kW), Indicates the unit The output power in time period t-1, that is, the output power in the previous time period t, Indicates the unit i The maximum output power change rate allowed is Indicates a time interval.

[0113] This formula specifies the upper and lower limits of power generation capacity and limits the power ramp rate. Due to the lag effect of the turbocharger, the diesel generator faces a stricter ramp rate limit (2 MW / min). This characteristic is expressed through the first-order delay transfer function. to model the time constant τ = 8 seconds.

[0114] The expression of the energy storage dynamic characteristic constraint is as follows:

[0115]

[0116] Where, Indicates energy storage The change in state of charge in time period t-1 is: Energy storage system j Charging efficiency, Energy storage system j The discharge efficiency, Energy storage system j Rated capacity, Energy storage system j The discharge power, Energy storage system j charging power.

[0117] In this formula, the efficiency factor (charging efficiency η ch =95%, discharge efficiency η dis =98%), modeling changes in state of charge (SOC), setting a safety margin to prevent lithium metal deposition, and limiting charge and discharge rates. Temporal coupling—where today's charging conditions affect tomorrow's discharge—requires multi-period optimization. This requires a second-generation non-dominated sorting genetic algorithm capable of handling 12,208 variables within a 24-hour timeframe.

[0118] The expression of renewable energy penetration constraint is as follows:

[0119]

[0120] This formula enables the optimizer to strike a balance between policy compliance and economic reality.

[0121] The expression of the coupling constraint between equipment and energy is as follows:

[0122]

[0123] Where, Indicates that the crane is at time t The power of M represents the number of rotating parts in the crane, and m is the index used to traverse the rotating parts, from 1 to M , representing different rotating parts, Indicates rotating parts m The angular velocity, Indicates rotating parts m The angular acceleration of represents the angle of the crane motor, 、 is the torque constant determined according to the manufacturer's specifications.

[0124] This constraint directly links mechanical operation to power demand, transforming the crane scheduling problem into a power trajectory optimization problem.

[0125] This disclosure causally integrates device physical characteristics with energy flows, enabling collaborative optimization of device operation and power dispatch. Furthermore, the three-objective (i.e., three objective functions) structure facilitates explicit trade-off analysis, facilitating decision-making for stakeholders.

[0126] The disclosed embodiments also validate the optimal configuration through extensive Monte Carlo simulations, incorporating probabilistic models for ship fuel delivery delays, random typhoon path evolution, and accelerated degradation of photovoltaic modules. Resilience assessment metrics include load recovery time during grid outages, the energy storage system's real-time state of charge (SOC), and the use of digital twin technology to replicate historical major failure events.

[0127] The effectiveness of this disclosure is demonstrated through a comparative analysis of three scenarios: Scenario S1 (a baseline scenario dominated by diesel power generation), Scenario S2 (a hybrid scenario of photovoltaic, wind, and energy storage), and Scenario S3 (an optimized system scenario). The three-objective Pareto frontier quantifies the inherent trade-offs between levelized cost of energy (LCOE), carbon intensity, and reliability.

[0128] Among them, scenario S3 has an advantage in the solution space, and its levelized energy cost and carbon dioxide emissions are lower than those of scenario S1.

[0129] This performance is due to time-of-day energy arbitrage: the lithium-ion battery storage system shifts 38% of the excess daytime PV power to the evening peak hours of crane demand, reducing reliance on LNG peaking units. However, the concave shape of the Pareto surface indicates that above 45 MWh of storage capacity, the benefits begin to diminish and the costs begin to increase, exceeding current carbon credit estimates.

[0130] Despite lower emissions, Scenario S2's levelized cost of energy increases. This (higher cost) stems from Scenario S2's static dispatch method, which fails to align PV generation with crane load troughs. Scenario S3 overcomes this limitation by intentionally delaying PV generation for a certain period of time.

[0131] The energy storage system played a dual role: lithium-ion batteries absorbed the load spikes caused by the cranes, while vanadium flow batteries provided typhoon resistance, enabling continuous power supply to critical loads during the historical blackouts. This hybrid energy storage strategy reduced reliance on diesel generation, primarily during the post-typhoon recovery phase, while the LNG peaking units operated selectively, avoiding the inefficient low-load operation prevalent in Scenario S1.

[0132] Scenario S2 results in multiple power outages due to reliance on a single type of energy storage, while Scenario S1 results in high fines due to fuel shortages, highlighting the disaster resilience of the optimized system.

[0133] The economic feasibility of Scenario S3, as measured by net present value (NPV), internal rate of return (IRR), and payback period, further validates the superiority of Scenario S3. The NPV of Scenario S3 is lower than that of Scenario S1. This economic advantage stems from load shifting incentives: energy storage arbitrage reduces the daily costs associated with rapid power demand and reduces outage costs.

[0134] Scenario S3 has a lower levelized cost of energy (LCOE) than hybrid microgrids and LNG-solar power systems. Its CO2 emissions intensity is similar to that of a system primarily based on offshore wind power, and its capital intensity is lower.

[0135] In summary, this disclosure establishes a transformative framework for optimizing port energy systems by integrating high-resolution load dynamics, the complementarity of multiple energy sources, and device-level transient response into a unified decision-making framework. Compared to a baseline scenario dominated by diesel generation, the optimized system lowers the levelized cost of energy and reduces CO2 emissions, while maintaining reliability during extreme weather events.

[0136] The analysis shows that the emission reductions come from load-shifting strategies that leverage the flexibility of crane scheduling and the thermal inertia of reefer containers, rather than relying on the addition of renewable energy. This challenges the prevailing paradigm of capacity-driven decarbonization and offers a scalable blueprint for ports with limited land resources.

[0137] Based on Figure 1 Based on the same principle as the method shown in , the present disclosure also provides a multi-energy coordination optimization device based on the port power load characteristics, Figure 7 This is a schematic diagram of a multi-energy coordination and optimization device based on port power load characteristics provided by an embodiment of the present disclosure. Figure 7 The multi-energy coordinated optimization 700 based on port power load characteristics may include:

[0138] The acquisition module 701 is used to obtain high-resolution power parameters of the load and power parameters supplied by renewable resources; the determination module 702 is used to match the high-resolution power parameters of the load and the power parameters supplied by renewable resources, determine the matching matrix of the power required by the load and the power supplied by renewable resources, and determine the power supply resource allocation strategy of the multi-energy storage system based on the matching matrix; the optimization module 703 is used to coordinate and optimize the power supply resource allocation strategy based on the objective function of the power supply resource allocation strategy and determine the dynamic optimization strategy.

[0139] In the embodiment of the present disclosure, the determination module 702 is used to extract load characteristics based on the high-resolution power parameters of the load, and extract renewable energy operation characteristics based on the renewable resource supply power parameters; determine a first interaction relationship between the load characteristics and the renewable energy operation, and determine a coupling model of the load and the renewable energy based on the first interaction relationship; use the coupling model to match the load characteristics and the renewable energy operation characteristics, and determine a matching matrix of the power required by the load and the power supplied by the renewable resources.

[0140] In the embodiment of the present disclosure, the determination module 702 is used to determine the load characteristic type based on the high-resolution power parameters of the load; determine the characteristic quantitative index of the load characteristic type and the renewable resource supply power parameter; determine tidal energy based on solar energy and wind energy of historical meteorological data; use the tidal energy as supplementary power supply during the low-power consumption period, match the supplementary power supply and the characteristic quantitative index through the matching matrix, and determine the power supply resource allocation strategy of the multi-energy storage system.

[0141] In the embodiment of the present disclosure, the determination module 702 is used to analyze the high-resolution power parameters of the load, determine the second interaction relationship between the operating conditions of the load, equipment energy consumption, and environmental factors and the equipment; and determine the load feature type based on the second interaction relationship.

[0142] In the embodiment of the present disclosure, the determination module 702 is configured to analyze the load characteristic type and determine a load pattern quantitative index, a load time regularity quantitative index, and a load power consumption fluctuation quantitative index;

[0143] Determine a third interaction relationship among the load pattern quantification index, the load time regularity quantification index, the load power consumption fluctuation quantification index and the renewable resource supply power parameter; analyze and quantify the third interaction relationship through spectral coherence to obtain a characteristic quantification index of the renewable resource.

[0144] In the present disclosure, the determination module 702 is used to obtain the determined load pattern quantitative index, load time regularity quantitative index, and load power consumption fluctuation quantitative index; based on the load pattern quantitative index, the load time regularity quantitative index, and the load power consumption fluctuation quantitative index, the lithium-ion battery configuration parameters, capacitor model parameters, and time parameters of using tidal energy as a supplementary power supply during the low-power period of the multi-energy energy storage system are determined; based on the lithium-ion battery configuration parameters, the capacitor model parameters, and the time parameters, the power supply resource allocation strategy of the multi-energy energy storage system is determined.

[0145] In the embodiment of the present disclosure, the optimization module 703 is used to dynamically adjust the Pareto front resolution based on the adaptive reference point and port size of the third-generation non-dominated sorting genetic algorithm, and determine the time weight based on the time discount effect of the third-generation non-dominated sorting genetic algorithm; determine the conflict mechanism between objectives of the objective function; weigh the conflict mechanism between objectives according to the adjusted Pareto front resolution and the time weight, dynamically optimize the objective function, and determine the dynamic optimization strategy of the power supply resource allocation strategy.

[0146] In the present disclosure, the objective function includes one or more of a levelized cost of electricity objective function, a carbon dioxide emission objective function, and a power shortage objective function.

[0147] Furthermore, the multi-objective optimization model also includes constraints;

[0148] The constraint conditions include one or more of power balance constraints, generator operation limitation constraints, energy storage dynamic characteristic constraints, renewable energy penetration constraints, and equipment and energy coupling constraints.

[0149] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A multi-energy coordination optimization method based on port power load characteristics is characterized by: The steps include: Obtain high-resolution power parameters of loads and power parameters supplied by renewable resources; Matching the load high-resolution power parameters and the renewable resource supply power parameters to determine a matching matrix between the load required power and the renewable resource supply power, and determining a power supply resource allocation strategy for the multi-energy storage system based on the matching matrix, including: extracting load characteristics based on the load high-resolution power parameters, and extracting renewable energy operation characteristics based on the renewable resource supply power parameters; determining a first interaction relationship between the load characteristic and the operation of the renewable energy source, and determining a coupling model of the load and the renewable energy source based on the first interaction relationship; Using the coupling model to match the load characteristics and the renewable energy operation characteristics, and determine a matching matrix between the power required by the load and the power supplied by the renewable resources; determining a load characteristic type based on the load high-resolution power parameter; Determining the load characteristic type and the characteristic quantitative index of the renewable resource supply power parameter; Determine tidal energy based on solar and wind energy from historical meteorological data; Using the tidal energy as supplementary power supply during the off-peak period, matching the supplementary power supply with the characteristic quantitative index through the matching matrix, and determining the power supply resource allocation strategy of the multi-energy storage system; Based on the objective function of the power supply resource allocation strategy, the power supply resource allocation strategy is coordinated and optimized to determine a dynamic optimization strategy, including: The resolution of the Pareto front is dynamically adjusted based on the adaptive reference point and port size of the third-generation non-dominated sorting genetic algorithm, and the time weight is determined based on the time discount effect of the third-generation non-dominated sorting genetic algorithm; Determine the conflict mechanism between objectives of the objective function; According to the adjusted Pareto front resolution and the time weight, the conflicting mechanisms among the objectives are weighed, the objective function is dynamically optimized, and a dynamic optimization strategy for the power supply resource allocation strategy is determined.

2. The multi-energy coordination optimization method based on port power load characteristics according to claim 1 is characterized in that: The determining of the load characteristic type based on the load high-resolution power parameter includes: Analyzing the high-resolution power parameters of the load to determine the operating condition of the load, device energy consumption, and a second interaction relationship between environmental factors and the device; According to the second interaction relationship, the load characteristic type is determined.

3. The multi-energy coordination optimization method based on port power load characteristics according to claim 1 is characterized in that: The determining of the load characteristic type and the characteristic quantitative index of the renewable resource supply power parameter includes: Analyze the load characteristic type to determine a load pattern quantitative index, a load time regularity quantitative index, and a load power consumption fluctuation quantitative index; Determining a third interaction relationship between the load pattern quantitative index, the load time regularity quantitative index, the load power consumption fluctuation quantitative index, and the renewable resource supply power parameter; The third interaction relationship is analyzed and quantified through spectral coherence to obtain characteristic quantitative indicators of renewable resources.

4. The multi-energy coordination optimization method based on port power load characteristics according to claim 1 is characterized in that: The method of using the tidal energy as a supplementary power supply during the off-peak period, matching the supplementary power supply with the characteristic quantitative index through the matching matrix, and determining a power supply resource allocation strategy for the multi-energy storage system includes: Obtaining quantitative indicators of determined load patterns, load time regularity, and load power fluctuation; Determining, based on the load pattern quantification index, the load time regularity quantification index, and the load power consumption fluctuation quantification index, the lithium-ion battery configuration parameters and capacitor model parameters of the multi-energy energy storage system, and the time parameters for using tidal energy as a supplementary power supply during the off-peak period; A power supply resource allocation strategy for a multi-energy storage system is determined based on the lithium-ion battery configuration parameters, the capacitor model parameters, and the time parameters.

5. The multi-energy coordination optimization method based on port power load characteristics according to claim 1 is characterized in that: The objective function includes one or more of a levelized cost of electricity objective function, a carbon dioxide emission objective function, and a power shortage objective function.

6. The multi-energy coordination optimization method based on port power load characteristics according to claim 1 is characterized in that: The objective function also includes constraints; The constraint conditions include one or more of power balance constraints, generator operation limitation constraints, energy storage dynamic characteristic constraints, renewable energy penetration constraints, and equipment and energy coupling constraints.

7. The multi-energy coordination and optimization device based on the port power load characteristics is characterized by: The multi-energy coordinated optimization method based on port power load characteristics according to any one of claims 1 to 6 is adopted, wherein the device comprises: An acquisition module is used to obtain high-resolution power parameters of the load and power parameters supplied by renewable resources; a determination module, configured to match the load high-resolution power parameters with the renewable resource supply power parameters, determine a matching matrix between the load required power and the renewable resource supply power, and determine a power supply resource allocation strategy for the multi-energy storage system based on the matching matrix; The optimization module is used to coordinate and optimize the power supply resource allocation strategy based on the objective function of the power supply resource allocation strategy and determine a dynamic optimization strategy.

Images