Port multi-energy collaborative dynamic optimization control system based on NSGA-III algorithm

Abstract

The present application relates to the technical field of multi-energy optimization control, in particular to a port multi-energy collaborative dynamic optimization control system based on NSGA-III algorithm, which comprises a target parameter definition module, a condition range adjustment module and a response weight calculation module. The target parameter definition module collects target parameters in the process of supplying energy utilization, matches the target function, and the condition range adjustment module adjusts the adaptability of the supply condition items. The supply state after the supply condition items are matched is calculated by the NSGA-III algorithm, and the corresponding response weight is calculated by the response weight calculation module. The supply energy of the matched supply condition items is selected through the response weight, and the ratio between the response weight and the actual response weight is fed back. The mutation value is identified, and the range of the supply condition items is adjusted with the mutation value. The adaptive range adjustment of the supply condition items is realized, the energy collaborative control effect is improved, and the energy supply adaptation degree is ensured.

Description

Technical Field

[0001] This invention relates to the field of multi-energy optimization control technology, and more specifically, to a port multi-energy collaborative dynamic optimization control system based on the NSGA-III algorithm. Background Technology

[0002] Port multi-energy synergistic dynamic optimization refers to the construction of a dynamic optimization model by integrating multiple energy types such as electricity, heat, and hydrogen energy, combined with the port's unique logistics needs (such as ship berthing and loading / unloading operations), in order to improve energy utilization efficiency, reduce operating costs, and achieve green and low-carbon operation.

[0003] In actual energy supply processes, the required energy supply intensity varies under different circumstances. For example, the energy supply for berthing or loading and unloading of large-tonnage ferries needs to be stable and continuous, while the energy supply intensity for berthing of small-tonnage ships is relatively low. At the same time, the energy demand required for different tasks also varies. Because the logistics demand in ports is complex and diverse, traditional port multi-energy collaborative dynamic control systems are difficult to achieve dynamic energy demand matching. In most cases, they will refer to historical supply records to adapt to energy demand, resulting in poor multi-energy collaborative dynamic optimization.

[0004] To address the aforementioned issues, there is an urgent need for a port multi-energy collaborative dynamic optimization control system based on the NSGA-III algorithm. Summary of the Invention

[0005] The purpose of this invention is to provide a port multi-energy collaborative dynamic optimization control system based on the NSGA-III algorithm. This system collects target parameters during the energy supply and utilization process through a target parameter definition module, matches the target function, and uses a condition range adjustment module to adaptively match supply condition items. The NSGA-III algorithm is then used to calculate the supply state after the matching of supply condition items. A response weight calculation module calculates the corresponding response weights, and the energy supply is selected based on these response weights. Simultaneously, the ratio between the response weights and the actual response weights is fed back in real time for variability identification. The range of supply condition items is then adjusted based on the variability values, achieving adaptive adjustment of the supply condition item range. This improves the subsequent energy collaborative control effect, ensures energy supply adaptability, and solves the problems mentioned in the background art, namely:

[0006] Traditional port multi-energy collaborative dynamic control systems are difficult to achieve dynamic energy demand matching. In most cases, they refer to historical supply records to adapt to energy demand, resulting in poor multi-energy collaborative dynamic optimization.

[0007] To achieve the above objectives, a port multi-energy collaborative dynamic optimization control system based on the NSGA-III algorithm is provided, including a port energy allocation module, a target parameter definition module, a condition range adjustment module, a response weight calculation module, and a target energy matching module.

[0008] The port energy distribution module controls the port energy distribution and converts the port's energy supply in real time.

[0009] Furthermore, for different energy supplies, different parameters are needed to assess whether the current energy demand is met. The target parameter definition module collects the target parameters in the energy utilization process and configures the corresponding target function to calculate the target parameter values.

[0010] The target parameters are calculated using the corresponding objective function. The corresponding supply status is calculated using the NSGA-III algorithm to obtain the final target parameter set and the same type of target parameter set. The clustering algorithm is then used to marginalize the target parameter values ​​in the same type of target parameter set.

[0011] If the percentage of values ​​in the edge clusters is less than the percentage threshold, then the target parameter values ​​in the edge clusters are removed, and the deviation values ​​are reclassified. and favorable value ;

[0012] If the proportion of values ​​in the edge cluster is not less than the value proportion threshold, then the target parameter values ​​in the edge cluster are retained.

[0013] Furthermore, since there are many different supply conditions in the port during the actual work process, it is necessary to collect supply conditions through the condition range adjustment module, and to make adaptive combinations of supply conditions. In conjunction with the target parameter definition module, the supply status after the combination of supply conditions is obtained. In conjunction with the response weight calculation module, the corresponding response weight is calculated based on the supply status after the combination of supply conditions, and the supply energy is matched through the response weight.

[0014] Simultaneously, the response weights are fed back to the condition range adjustment module to adjust the matching range of various supply condition items in real time. By defining the unit adjustment range for each supply condition item and defining the adjustment priority for each supply condition item, the adjustment range is adjusted by one unit at a time, and the corresponding target parameter sets of the same type are recalculated. And response weights enable adaptive adjustment of the project scope based on supply conditions, improve the effectiveness of subsequent energy coordination control, and ensure energy supply adaptability.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0016] In this port multi-energy collaborative dynamic optimization control system based on the NSGA-III algorithm, target parameters in the energy supply and utilization process are collected through a target parameter definition module, and the target function is matched. The condition range adjustment module adaptively matches the supply condition items, and the NSGA-III algorithm is used to calculate the supply state after the matching of supply condition items. The response weight calculation module calculates the corresponding response weight, and the energy supply is selected and matched according to the matched supply condition items. At the same time, the ratio between the response weight and the actual response weight is fed back in real time to identify the variation value. The range of supply condition items is adjusted according to the variation value to achieve adaptive adjustment of the range of supply condition items, thereby improving the subsequent energy collaborative control effect and ensuring the adaptability of energy supply. Attached Figure Description

[0017] Figure 1 This is a block diagram of the overall structure of the present invention.

[0018] The meanings of the labels in the diagram are as follows:

[0019] 10. Port Energy Distribution Module;

[0020] 20. Target parameter definition module;

[0021] 30. Condition range adjustment module;

[0022] 40. Response weight calculation module;

[0023] 50. Target Energy Matching Module. Detailed Implementation

[0024] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] Please see Figure 1 As shown, a port multi-energy collaborative dynamic optimization control system based on the NSGA-III algorithm is provided, including a port energy allocation module 10, a target parameter definition module 20, a condition range adjustment module 30, a response weight calculation module 40, and a target energy matching module 50.

[0026] Port energy distribution module 10 is used to control port energy distribution and switch port energy supply in real time;

[0027] The target parameter definition module 20 collects target parameters during the energy supply and utilization process, matches the target function, and uses the NSGA-III algorithm to calculate the corresponding supply state.

[0028] The condition range adjustment module 30 is used to collect supply condition items and perform adaptive matching of supply condition items, and in conjunction with the target parameter definition module 20, obtain the supply status after matching supply condition items;

[0029] The response weight calculation module 40 calculates the corresponding response weight based on the supply status after the supply condition items are matched, and feeds the response weight back to the condition range adjustment module 30 to adjust the matching range of each supply condition item in real time.

[0030] The target energy matching module 50 obtains the response weights after adjusting the matching range of supply condition items, and matches supply energy to supply condition items under different matching range states according to the response weights.

[0031] The details are as follows:

[0032] Firstly, to meet diverse logistics needs, ports require multiple energy supplies. To adapt to these needs, the port energy distribution module 10 controls the port's energy allocation, switching between various energy sources in real time, such as thermal power, nuclear power, and green energy. Different energy sources have different advantages and disadvantages. Thermal and nuclear power offer greater stability and are suitable for conventional loading and unloading equipment, lighting systems, office area power supply, and refrigerated container yards. They require a stable power grid and are suitable for fixed facility operations, but have higher carbon emissions.

[0033] For green energy, it is suitable for auxiliary power supply scenarios, such as monitoring facilities.

[0034] Furthermore, for different energy supplies, different parameters are needed to assess whether the current energy demand is met. The target parameter definition module 20 collects target parameters during the energy utilization process. In this scheme, the corresponding target parameters include utilization efficiency, equipment wear and tear, operating costs, and environmental friendliness. Objective functions are configured for these target parameters. The objective function for operating costs is configured as the product of the consumption of various energy sources and their corresponding unit prices, taking into account maintenance costs and equipment depreciation costs to determine the total operating cost. The corresponding objective function algorithm formula is as follows:

[0035] ;

[0036] in, Total operating cost, For the i-th type of energy consumption, Let i be the unit price of energy of type i. D represents maintenance costs, and D represents equipment depreciation costs.

[0037] The objective function for energy utilization efficiency is configured to calculate and reflect energy utilization efficiency through the percentage relationship between effective output energy and total input energy. The corresponding objective function algorithm formula is as follows:

[0038] ;

[0039] in, Energy efficiency percentage To effectively output energy, such as mechanical energy and thermal energy, The total input energy is typically fuel or electricity;

[0040] The objective function for equipment loss is configured to evaluate the equipment loss rate based on the ratio of actual operating time to the total design life, and by taking into account environmental factors, average load rate, and load sensitivity index. The corresponding objective function algorithm formula is as follows:

[0041] ;

[0042] in, Equipment wear rate This refers to the actual running time. For the total design lifespan, This is an environmental factor, generally influenced by temperature and humidity. Average load factor This is the load sensitivity index, typically ranging from 1.5 to 2.5.

[0043] The environmental impact objective function is configured to measure total carbon emissions by accumulating the product of various fuel consumptions and their corresponding emission factors. The corresponding objective function algorithm formula is as follows:

[0044] ;

[0045] in Total carbon emissions For the k-th type of fuel consumption, The emission factor for fuel type k, for example, the emission factor for diesel is 2.68 kg. L represents liters, and the natural gas emission factor is 2.75 kg. .

[0046] Each objective parameter is calculated using its corresponding objective function, and the corresponding supply state is calculated using the NSGA-III algorithm. The specific algorithm steps are as follows:

[0047] First, define different supply condition ranges, such as berthing time ranges. Collect the values ​​of various target parameters for different energy supplies under the same supply condition range, and mark the favorable trends of each target parameter. For example, for operating costs, the smaller the value, the more favorable it is for port operation demand. Therefore, the corresponding favorable trend is that the smaller the value, the more favorable it is. Then, aggregate and process the collected target parameter values ​​to generate a target parameter set. ,in This is the current initial supply condition range. This represents the operating cost figure. To utilize efficiency values, This represents equipment wear and tear. To calculate the total carbon emissions, we statistically analyze the target parameter sets for the same energy supply under multiple sets of identical initial supply conditions. Obtain a set of target parameters of the same type. For example, a similar set of target parameters for operating costs. for ,in to For operating cost values ​​collected at different time stages within the same energy supply under the same initial supply conditions, a favorable value is obtained based on the favorable trend of the current target parameters. and deviation value For example, a similar set of target parameters for operating costs. for , to The value gradually increases, while the favorable trend for operating costs is that the smaller the value, the more advantageous it is; therefore, the corresponding advantageous value... Then it is Deviation value Then it is ;

[0048] Furthermore, in order to improve the same type of target parameter set To improve accuracy, a clustering algorithm is used to analyze similar target parameter sets. To perform clustering, first start with a set of similar target parameters. Select one or more target parameter values ​​(this is based on a set of similar target parameters). The numerical values ​​are defined, and the more numerical values ​​there are, the more selections are adjusted accordingly. These are used as initial cluster centers to establish cluster sets, and sets of similar target parameters are calculated. The distance between each target parameter value and the initial cluster center is used to classify similar target parameter sets. Each target parameter value is assigned to the nearest cluster, completing the first clustering process. Then, the average target parameter value across all clusters is calculated and used as the new cluster center. New clusters are established based on these centers, and the distance between each cluster's target parameter value and its corresponding cluster center is calculated. The target parameter values ​​are then assigned to the nearest clusters. This process is repeated until the values ​​in all clusters remain unchanged. At this point, the edge clusters (i.e., those primarily used to store deviation values) are selected. and favorable value The percentage of values ​​in the edge cluster is calculated as: (Number of target parameters in the edge cluster / Number of similar target parameter sets). The total number of target parameters is used to establish a threshold for the percentage of values, and a judgment is made accordingly.

[0049] If the percentage of values ​​in the edge clusters is less than the percentage threshold, then the target parameter values ​​in the edge clusters are removed, and the deviation values ​​are reclassified. and favorable value ;

[0050] If the proportion of values ​​in the edge cluster is not less than the value proportion threshold, then the target parameter values ​​in the edge cluster are retained.

[0051] Furthermore, since the port supply conditions are diverse in the actual work process, it is necessary to collect the supply condition items through the condition range adjustment module 30, and to perform adaptive matching of the supply condition items. In conjunction with the target parameter definition module 20, the supply status after the matching of supply condition items is obtained. The specific method is as follows:

[0052] First, obtain the supply condition items for the current port logistics demand, such as the combination of operation content and berthing time. Calculate the numerical range of the current supply condition items, and based on the numerical range, obtain the set of similar target parameters for different supply condition items under different energy supply conditions. Select the target parameter set of each type The corresponding favorable value Establish a favorable set of data A supply condition item corresponds to a favorable set of values. .

[0053] After completing the supply status calculation following the matching of supply conditions, the response weight calculation module 40 calculates the corresponding response weight based on the supply status after the matching of supply conditions. The specific calculation method is as follows:

[0054] First, define the weighting coefficients for each objective parameter, and then calculate the favorable value set after matching the supply conditions with the project. The response weights in the algorithm are calculated using the following formula:

[0055] ;

[0056] in As the response weight for one of the supply condition items, The weighting factor for operating costs, For a favorable set of values The operating cost advantage value in the middle, This is a weighting factor for energy efficiency. For a favorable set of values Regarding the beneficial value of utilization efficiency, it's worth noting that to eliminate differences in units and dimensions, the utilization efficiency values ​​need to be normalized. This is done by normalizing the utilization efficiency using a minimum-maximum value, yielding the corresponding beneficial value. Similarly, the actual utilization efficiency values ​​below also need to be normalized using a minimum-maximum value. This is the weighting coefficient for equipment losses. For a favorable set of values The beneficial value of equipment loss in the middle, As a weighting factor for environmental friendliness, For a favorable set of values The environmental benefits of this product;

[0057] The supply conditions items are calculated sequentially according to the above algorithm. And calculate the sum of response weights after matching supply conditions items.

[0058] Furthermore, to improve the accuracy of response weight matching, the response weight needs to be fed back to the condition range adjustment module 30 to adjust the matching range of various supply condition items in real time. The specific adjustment method is as follows:

[0059] First, the values ​​of various target parameters during the actual working process are collected and marked as actual target parameter values. Based on the weight coefficients of each target parameter, the corresponding actual response weights are calculated. The specific algorithm is as follows:

[0060] ;

[0061] in For actual response weights, For actual operating costs, For actual utilization efficiency, This represents actual equipment wear and tear. For practical environmental protection;

[0062] The actual response weight is then calculated. The ratio of the sum of response weights when combined with supply condition items is denoted as the weight ratio. Set ratio thresholds And calculate the comparison;

[0063] When weight ratio Exceeding the ratio threshold When this happens, the sum of the response weights after matching the supply condition items is marked as the variation value;

[0064] When weight ratio Not exceeding the ratio threshold In this case, the sum of the response weights after matching the supply conditions items is retained;

[0065] Obtain the amount of variation for the statistical time period and determine the threshold for the amount of variation;

[0066] If the amount of variation within the statistical time period does not exceed the value of variation threshold, no adjustment to the supply condition range is required.

[0067] If the amount of variation within the statistical time period exceeds the threshold for the amount of variation, the range of supply conditions in the current supply condition item combination needs to be adjusted.

[0068] During the adjustment process, it is necessary to determine the unit adjustment range based on each supply condition item, define the specific values ​​of each supply condition item, and define the adjustment priority of each supply condition item. For example, when work content is paired with docking time, since work content is dominant, its adjustment priority is higher than docking time. That is, the range of work content needs to be adjusted first, adjusting one unit range at a time, and recalculating the corresponding set of similar target parameters. And response weights enable adaptive adjustment of the project scope based on supply conditions, improve the effectiveness of subsequent energy coordination control, and ensure energy supply adaptability.

[0069] Finally, in conjunction with the target energy matching module 50, the response weights of the adjusted supply condition items are obtained. Based on the response weights, supply energy is matched for supply condition items under different matching ranges. The numerical range of different response weights corresponds to the supply energy under different supply condition items. During matching, matching is performed based on the division results. That is, the response weight of the corresponding supply energy is calculated based on the range of the currently matched supply condition items, and the supply energy with the highest response weight is selected as the matched supply energy for the currently matched supply condition item.

[0070] This invention collects target parameters during the energy supply and utilization process through a target parameter definition module 20, matches the target function, and uses a condition range adjustment module 30 to adaptively match supply condition items. It then uses the NSGA-III algorithm to calculate the supply status after the matching of supply condition items and a response weight calculation module 40 to calculate the corresponding response weights. These response weights are used to select and match the energy supply for the matched supply condition items, while simultaneously providing real-time feedback on the ratio between the response weights and the actual response weights to obtain actual matching differences. This allows for the identification of variability values, and the adjustment of the supply condition item range based on these variability values. This achieves adaptive adjustment of the supply condition item range, improving subsequent energy coordination control and ensuring energy supply adaptability.

[0071] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A port multi-energy collaborative dynamic optimization control system based on the NSGA-III algorithm, characterized in that: It includes a port energy allocation module (10), a target parameter definition module (20), a condition range adjustment module (30), a response weight calculation module (40), and a target energy matching module (50). The port energy distribution module (10) is used to control the port energy distribution and switch the port's energy supply in real time; The target parameter definition module (20) collects the target parameters in the energy supply utilization process, matches the target function, and uses the NSGA-III algorithm to calculate the corresponding supply state. The condition range adjustment module (30) is used to collect supply condition items and perform adaptive matching of supply condition items, and cooperate with the target parameter definition module (20) to obtain the supply status after matching of supply condition items; The response weight calculation module (40) calculates the corresponding response weight based on the supply status after the supply condition items are matched, and feeds the response weight back to the condition range adjustment module (30) to adjust the matching range of each supply condition item in real time. The target energy matching module (50) obtains the response weight after adjusting the matching range of the supply condition items, and matches the supply energy for the supply condition items under different matching range states according to the response weight; The target parameters collected by the target parameter definition module (20) include utilization efficiency, equipment wear and tear, operating cost and environmental friendliness; The method for calculating the corresponding supply state using the NSGA-III algorithm in the target parameter definition module (20) includes the following steps: S201. Define different supply condition ranges, collect the values ​​of various target parameters for different energy supplies under the same supply condition range, and mark the favorable trends of each target parameter; S202、And the collected target parameter value is set processing, generating a target parameter set ; S203, count the target parameter set of the same energy supply under the same initial supply condition range of multiple groups , obtain the same target parameter set ; S204, obtaining a favorable value according to the favorable trend of the current target parameter and the deviation value The method of S203 includes the following steps: The method includes the following steps: S2031, From a similar set of target parameters Select one or more target parameter values ​​as the initial cluster centers; S2032, establish a cluster set with the initial cluster center, calculate the same target parameter set the distance between each target parameter value and the initial cluster center; S2033, Use similar target parameter sets The values ​​of each target parameter are assigned to the nearest cluster, completing the first clustering process. S2034. Calculate the average target parameter value in each cluster set, use the average target parameter value as the cluster center of the new round, and establish a new cluster set based on the divided cluster centers. Continue to calculate the distance between the target parameter value in each cluster set and each cluster center, and assign the target parameter value in each cluster set to the nearest cluster set. S2035. Repeat step S2034 until the values ​​in all clusters do not change. S2036, select the edge cluster set, obtain the value proportion in the edge cluster set = the number of target parameters in the edge cluster set / the total amount of target parameters in the same type target parameter set , establish the value proportion threshold value, and determine; If the percentage of values ​​in the edge clusters is less than the percentage threshold, then the target parameter values ​​in the edge clusters are removed, and the deviation values ​​are reclassified. and favorable value ; If the proportion of values ​​in the edge cluster is not less than the value proportion threshold, then the target parameter values ​​in the edge cluster are retained.

2. The port multi-energy coordinated dynamic optimization control system based on the NSGA-III algorithm of claim 1, characterized in that: The objective function for the operating cost is configured to sum the product of the consumption of various energy sources and their corresponding unit prices, and to include maintenance costs and equipment depreciation costs, in order to determine the total operating cost.

3. The NSGA-III algorithm-based port multi-energy coordinated dynamic optimization control system according to claim 1, characterized in that: The objective function for the utilization efficiency is configured to calculate and reflect energy utilization efficiency by using the percentage relationship between effective output energy and total input energy.

4. The NSGA-III algorithm-based port multi-energy coordinated dynamic optimization control system according to claim 1, characterized in that: The objective function for equipment loss is configured to assess the equipment loss rate based on the ratio of actual operating time to the total design life, and by taking into account environmental factors, average load rate, and load sensitivity index.

5. The NSGA-III algorithm-based port multi-energy coordinated dynamic optimization control system according to claim 1, characterized in that: The environmental impact objective function is configured to measure total carbon emissions by accumulating the product of various fuel consumptions and their corresponding emission factors.

6. The port multi-energy collaborative dynamic optimization control system based on the NSGA-III algorithm according to claim 1, characterized in that: The method for adaptively matching supply conditions in the condition range adjustment module (30) includes the following steps: S301. Obtain the current port logistics demand supply condition item combination and calculate the numerical range of the current combination supply condition item. S302, obtaining, according to the numerical range, each set of similar target parameters of different supply condition items under different energy supply conditions ; S303, selecting each same kind target parameter set The corresponding favorable value , establishing a favorable value set One supply condition item corresponds to one favorable value set .

7. The port multi-energy collaborative dynamic optimization control system based on the NSGA-III algorithm according to claim 1, characterized in that: The method for adjusting the matching range of various supply condition items in the response weight calculation module (40) includes the following steps: S401、Define the weight coefficient of each target parameter, calculate the favorable value set after the supply condition item combination The response weight in the formula is as follows: ； in As the response weight for one of the supply condition items, The weighting factor for operating costs, For a favorable set of values The operating cost advantage value in the middle, This is a weighting factor for energy efficiency. For a favorable set of values The utilization efficiency value in This is the weighting coefficient for equipment losses. For a favorable set of values The beneficial value of equipment loss in the middle, As a weighting factor for environmental friendliness, For a favorable set of values The environmental benefits of this product; S402、According to the above algorithm, the supply condition items are calculated in turn and the sum of the response weights of the supply condition item combinations is calculated. S403. Collect the values ​​of various target parameters during the actual working process, mark them as actual target parameter values, and calculate the corresponding actual response weights based on the weight coefficients of each target parameter. The specific algorithm is as follows: ； wherein is the actual response weight, is the actual operating cost, is the actual utilization efficiency, is the actual equipment loss, is the actual environmental friendliness; S404. Calculate the actual response weight. The ratio of the sum of response weights when combined with supply condition items is denoted as the weight ratio. Set ratio thresholds And calculate the comparison; When the weight ratio exceeds the ratio threshold , then the sum of the response weights after the supply condition item is matched is marked as a variation value; When the weight ratio The sum of the response weights after the supply condition item collocation is retained when the weight ratio is not exceeded, S405. Obtain the amount of variation in the statistical time period and set the threshold for the amount of variation; If the amount of variation within the statistical time period does not exceed the value of variation threshold, no adjustment to the supply condition range is required. If the amount of variation within the statistical time period exceeds the threshold for the amount of variation, the range of supply conditions in the current supply condition item combination needs to be adjusted. S406. Determine the scope of unit adjustments based on each supply condition item, and define the adjustment priority for each supply condition item; S407. Adjust the adjustment range by one unit at a time, and recalculate the corresponding set of similar target parameters. And response weights.

Images