Port green strategy intelligent optimization method and device based on elastic network regression weighting, computer equipment and readable storage medium

Abstract

The invention discloses a port green strategy intelligent optimization method and device based on elastic network regression weighting, computer equipment and a readable storage medium, and the method comprises the steps: firstly obtaining a plurality of first-level port green correlation indexes, including energy conservation, low carbon, port efficiency and the like, of a target port; and then calculating a port greening strategy corresponding to each index and a full life cycle cost-benefit result. Then, an elastic network regression model is constructed based on the indexes and the port greening degree, and weight information of all the indexes is obtained; and finally, a green strategy combination optimal solution is determined through a genetic algorithm in combination with a cost-benefit result and weight information, intelligent optimization of a port greening strategy is realized, and the port greening level and the sustainable development capability are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of port management. Specifically, it relates to an intelligent optimization method, device, computer equipment and readable storage medium for port green strategies based on elastic net regression weighting. Background Art

[0002] Currently, most port greening evaluation methods are qualitative or semi-quantitative analyses, lacking a systematic quantitative index system and making it difficult to comprehensively and accurately measure the port greening level. In terms of cost-benefit evaluation, the quantitative analysis is insufficient, and there is a lack of an accurate quantitative model for ecological and social benefits. There are also deficiencies in the optimization and recommendation methods for greening improvement strategies, lacking a targeted strategy optimization model and comprehensive benefit evaluation. The existing technologies cannot fully meet the needs of port sustainable development. Summary of the Invention

[0003] The purpose of the present invention is to provide an intelligent optimization method, device, computer equipment and readable storage medium for port green strategies based on elastic net regression weighting.

[0004] In a first aspect, an embodiment of the present invention provides an intelligent optimization method for port green strategies based on elastic net regression weighting, which is characterized by including: obtaining a plurality of first-level port green correlation indicators corresponding to a target port, where the plurality of first-level port green correlation indicators include energy conservation and low-carbon indicators, port efficiency indicators, port pollution prevention and control indicators, port resource recycling indicators, and ecological friendliness indicators; respectively calculating the port greening strategies corresponding to each of the first-level port green correlation indicators, and determining the life-cycle cost-benefit results of each of the port greening strategies; constructing an elastic net regression model based on the plurality of first-level port green correlation indicators and the port greening degree, and calculating the weight information corresponding to each of the first-level port green correlation indicators; and determining the optimal solution of the green strategy combination for the target port through a genetic algorithm based on the life-cycle cost-benefit results and the weight information.

[0005] Second aspect, an embodiment of the present invention provides a smart optimization device for port green strategies based on elastic net regression weighting, including: an acquisition module, configured to acquire a plurality of first-level port green correlation indicators corresponding to a target port, the plurality of first-level port green correlation indicators including energy conservation and low-carbon indicators, port efficiency indicators, port pollution prevention and control indicators, port resource recycling indicators, and ecological friendliness indicators; a calculation module, configured to calculate respective port greening strategies corresponding to each of the first-level port green correlation indicators, and determine the life-cycle cost-benefit results of each of the port greening strategies; an optimization module, configured to construct an elastic net regression model based on the plurality of first-level port green correlation indicators and the port greening degree, calculate the weight information corresponding to each of the first-level port green correlation indicators; and determine an optimal solution of the green strategy combination for the target port through a genetic algorithm based on the life-cycle cost-benefit results and the weight information. Third aspect, an embodiment of the present invention provides a computer device, the computer device includes a processor and a non-volatile memory storing computer instructions, when the computer instructions are executed by the processor, the computer device executes the method described in the first aspect. Fourth aspect, an embodiment of the present invention provides a readable storage medium, the readable storage medium includes a computer program, when the computer program runs, it controls the computer device where the readable storage medium is located to execute the method described in the first aspect.

[0006] Compared with the prior art, the beneficial effects provided by the present invention include: adopting a smart optimization method, device, computer device, and readable storage medium for port green strategies based on elastic net regression weighting disclosed in the present invention, by acquiring a plurality of first-level port green correlation indicators including energy conservation and low-carbon, port efficiency, etc. for a target port, then calculating the port greening strategies corresponding to each indicator and the life-cycle cost-benefit results. Then, an elastic net regression model is constructed based on these indicators and the port greening degree to obtain the weight information of each indicator. Finally, combining the cost-benefit results and the weight information, an optimal solution of the green strategy combination is determined through a genetic algorithm, realizing the smart optimization of port green strategies and improving the port greening level and sustainable development ability. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention, and thus should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0008] Figure 1 It is a schematic flowchart of the steps of the smart optimization method for port green strategies based on elastic net regression weighting provided by the embodiment of the present invention; Figure 2 Schematic diagram of the framework of the intelligent optimization system for port green strategies based on elastic net regression weighting provided by an embodiment of the present invention; Figure 3 Block diagram of the structure of the intelligent optimization device for port green strategies based on elastic net regression weighting provided by an embodiment of the present invention; Figure 4 Block diagram of the structure of the computer device provided by an embodiment of the present invention. Specific embodiments

[0009] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Generally, the components of the embodiments of the present invention described and illustrated in the drawings here can be arranged and designed in various different configurations.

[0010] The following will describe in detail the specific embodiments of the present invention with reference to the accompanying drawings.

[0011] To solve the technical problems in the foregoing background art, Figure 1 Schematic diagram of the process of the intelligent optimization method for port green strategies based on elastic net regression weighting provided by an embodiment of the present disclosure. The intelligent optimization method for port green strategies based on elastic net regression weighting will be introduced in detail below.

[0012] Step S201: Obtain a plurality of first-level port green correlation indicators corresponding to the target port, where the plurality of first-level port green correlation indicators include energy conservation and low-carbon indicators, port efficiency indicators, port pollution prevention and control indicators, port resource recycling indicators, and ecological friendliness indicators; Step S202: Calculate the port greening strategies corresponding to each of the first-level port green correlation indicators respectively, and determine the life-cycle cost-benefit results of each of the port greening strategies; Step S203: Construct an elastic net regression model based on the plurality of first-level port green correlation indicators and the port greening degree, and calculate the weight information corresponding to each of the first-level port green correlation indicators; Step S204: Based on the life-cycle cost-benefit results and the weight information, determine the optimal solution of the green strategy combination for the target port through a genetic algorithm.

[0013] In an embodiment of the present invention, exemplarily, the server begins to optimize the greening strategy of XX Port. First, the server obtains multiple primary port green correlation indicators of XX Port, covering energy conservation and low carbon, port efficiency, port pollution prevention, port resource recycling, and ecological friendliness indicators. In terms of the energy conservation and low carbon indicators, the proportion of low-carbon energy, the proportion of renewable energy terminals, the energy consumption intensity, shore power-related data, and the number of new energy supply points are counted. For the port efficiency indicators, the access rate of the inbound railway line, the proportion of non-road collection and distribution, and the ship stay time in the port are counted. Among the port pollution prevention indicators, multiple proportions such as the municipal pipe network coverage rate and the number of facilities are calculated. In the port resource recycling indicators, proportions such as the cold energy utilization rate are calculated. In terms of the ecological friendliness indicators, the number of ecological structure projects adopted, etc., is counted. Then, for each primary indicator, the corresponding port greening strategy and the life cycle cost-benefit results are calculated. For the energy conservation and low carbon strategy, clean energy is adopted and facilities are built. The costs cover aspects such as equipment procurement. The benefits are the savings in energy costs and the reduction of pollutants. For the port efficiency improvement strategy, multimodal transport is implemented and the layout is optimized. The costs involve infrastructure transformation, etc. The benefits include the increase in throughput, etc. For the resource recycling strategy, the reuse of existing facilities, etc., is carried out. The costs include dredging material treatment costs, etc. The benefits include landfill cost savings, etc. For the pollution prevention strategy, the sewage pipe network is connected, etc. The costs include construction, operation and maintenance costs, etc. The benefits include water quality improvement, etc. For the ecological friendliness strategy, ecological structure projects, etc., are built. The costs include seedling procurement, etc. The benefits include the restoration of fishery resources, etc. Subsequently, the server constructs an elastic net regression model based on the obtained indicators and the port greening degree. Taking the port greening degree as the target variable and 22 sub-indicators as the input, through calculation, the relationship between the indicators and the greening degree is comprehensively considered, and L1 and L2 regularization are used to control the model complexity and fitting degree, and the problem of multicollinearity among the indicators is processed. Finally, the corresponding weight information of each primary indicator is obtained, reflecting its contribution to the port greening degree. For example, the weight of the energy conservation and low carbon indicator may be relatively high. Finally, the server uses the genetic algorithm to determine the optimal solution of the green strategy combination. First, an initial solution space is randomly generated, such as combinations of different clean energy promotion ratios, etc. Then, the inverse function of the objective function is used as the fitness function. This objective function combines cost-benefit and weights. The higher the fitness, the higher the greening benefit under the same cost. Then, crossover and mutation operations are carried out, such as crossing different focused strategies to generate new combinations, and a small amount of mutation is introduced to prevent local optimality. In the iteration, natural selection is simulated, the poor-performing combinations are eliminated, and the high-fitness ones are retained. When the maximum iteration number is reached or the fitness converges, the optimal solution is output to ensure that XX Port achieves the maximum greening benefit under the limited cost, such as adopting a strategy combination of clean energy and optimized resource recycling to improve environmental protection and economic benefits.

[0014] In a possible implementation manner, the obtaining of multiple primary port green correlation indicators corresponding to the target port may be executed through the following examples.

[0015] Take the proportion of low-carbon energy, the proportion of terminals using renewable energy, energy consumption intensity, the proportion of berths with shore power available, the utilization rate of shore power, and other energy supply capabilities as multiple secondary port green correlation indicators of the energy-saving and low-carbon indicators; take the access rate of inbound railway lines, the proportion of non-road collection and distribution, and the stay time of ships in port as multiple secondary port green correlation indicators of the port efficiency indicators; take the coverage rate of municipal pipe networks, the closure rate of dry bulk cargo yards, the coverage rate of oil and gas recovery, the compliance rate of port boundary noise, the classification collection and disposal rate of port solid waste, the coverage rate of ship pollutant receiving facilities, and the construction of ballast water receiving facilities as multiple secondary port green correlation indicators of the port pollution prevention indicators; take the cold energy utilization rate, the comprehensive utilization rate of dredged soil, and the utilization of unconventional water resources as multiple secondary port green correlation indicators of the port resource recycling indicators; take the number of projects using ecological structures, the number of ecological restoration and rehabilitation projects, and the special protection plan for important organisms as multiple secondary port green correlation indicators of the ecological friendliness indicators; based on the multiple secondary port green correlation indicators, construct multiple primary port green correlation indicators corresponding to the target port.

[0016] In an embodiment of the present invention, by way of example, it is assumed that the target port is "Port A", and the server performs the task of obtaining multiple primary port green correlation indicators. For the energy conservation and low-carbon indicators, the server takes the proportion of low-carbon energy, the proportion of terminals applying renewable energy, the energy consumption intensity, the proportion of berths with shore power supply, the shore power utilization rate, and other energy supply capabilities as its multiple secondary port green correlation indicators. In Port A, the server calculates the proportion of low-carbon energy such as electricity and LNG in the total energy consumption to determine the proportion of low-carbon energy. It checks the number of terminals using renewable energy such as solar energy and wind energy, and calculates the proportion of terminals applying renewable energy compared with the total number of terminals. The energy consumption intensity is calculated by the ratio of the total energy consumption of the port to the throughput. The server calculates the proportion of the number of berths with shore power supply capacity to the total number of berths to obtain the proportion of berths with shore power supply. The proportion of the number of ships using shore power to the number of ships entering the port is calculated to determine the shore power utilization rate. At the same time, the number of supply points of new ship energy such as ammonia and hydrogen is counted as other energy supply capabilities. Based on these secondary indicators, the energy conservation and low-carbon indicators of Port A are constructed. In terms of port efficiency indicators, the server takes the access rate of inbound railway lines, the proportion of non-road collection and distribution, and the ship stay time in port as multiple secondary port green correlation indicators. The server calculates the number of operating areas with dedicated inbound railway lines built and calculates the access rate of inbound railway lines compared with the total number of operating areas. The proportion of non-road collection and distribution volume such as waterway and railway in the total collection and distribution volume is calculated to obtain the proportion of non-road collection and distribution. The sum of the waiting time at anchor and the loading and unloading time of the ship is recorded as the ship stay time in port. Based on these secondary indicators, the port efficiency indicators of Port A are constructed. For the port pollution prevention and control indicators, the server takes the coverage rate of municipal pipe networks, the closure rate of dry bulk cargo yards, the coverage rate of oil and gas recovery, the compliance rate of port boundary noise, the classification and disposal rate of port solid waste, the coverage rate of ship pollutant receiving facilities, and the construction of ballast water receiving facilities as multiple secondary port green correlation indicators. In Port A, the server calculates the proportion of the number of terminals that can be connected to the municipal sewage pipe network to the total number of terminals to obtain the coverage rate of municipal pipe networks. The proportion of the closed area of the dry bulk cargo yard to the total area is counted to determine the closure rate of the dry bulk cargo yard. The proportion of the number of terminals with oil and gas recovery capabilities to the total number of liquid bulk terminals is counted to obtain the coverage rate of oil and gas recovery. The proportion of the number of terminals meeting the noise standard to the total number of terminals is calculated to obtain the compliance rate of port boundary noise. The proportion of the classified and disposed solid waste volume to the total solid waste volume is counted to determine the classification and disposal rate of port solid waste. The proportion of the number of berths with pollutant receiving facilities to the total number of berths is counted to obtain the coverage rate of ship pollutant receiving facilities. The number of fixed and mobile ballast water receiving facilities is counted as the construction of ballast water receiving facilities. Based on these secondary indicators, the port pollution prevention and control indicators are constructed. For the port resource recycling indicators, the server takes the cold energy utilization rate, the comprehensive utilization rate of dredged soil, and the utilization of unconventional water resources as multiple secondary port green correlation indicators.At the LNG terminal in Port A, the server calculates the ratio of the actually utilized cold energy to the total generated cold energy to obtain the cold energy utilization rate. The server also calculates the ratio of the volume of reutilized dredged soil from the terminal and waterway to the total volume of dredged soil to obtain the comprehensive utilization rate of dredged soil. The server further calculates the ratio of the number of terminals using unconventional water resources such as production wastewater and ship sewage to the total number of terminals to determine the utilization of unconventional water resources. Based on these secondary indicators, a port resource reuse indicator is constructed. In terms of the eco-friendly indicator, the server uses the number of projects with ecological structures, the number of ecological restoration and rehabilitation projects, and the special protection plans for important organisms as multiple secondary port green correlation indicators. The server counts the number of terminals and protection projects with ecological structures during the planning period of Port A, counts the number of ecological restoration and rehabilitation projects led by the port or funded by enterprises, and counts the number of special protection plans for important species that may be affected by the port. Based on these secondary indicators, an eco-friendly indicator is constructed, and finally, the construction of multiple primary port green correlation indicators for the target port is completed.

[0017] In the embodiment of the present invention, the respective port greening strategies corresponding to each of the primary port green correlation indicators are calculated separately, and the life-cycle cost-benefit results of each of the port greening strategies are determined, which can be implemented through the following examples.

[0018] Calculate the energy facility cost, annual energy consumption cost, annual operation and maintenance cost, and equipment depreciation cost as the costs of the energy-saving and low-carbon strategies corresponding to the energy-saving and low-carbon indicator, and calculate the energy cost savings and reduction in air pollutants as the benefits of the energy-saving and low-carbon strategies corresponding to the energy-saving and low-carbon indicator; calculate the intermodal internal transfer cost as the cost of the port efficiency indicator strategy corresponding to the port efficiency indicator, and calculate the increased throughput revenue, optimized land use revenue, and increased internal transfer efficiency revenue as the benefits of the port efficiency indicator strategy corresponding to the port efficiency indicator; calculate the domestic sewage pipe network connection cost, waste sorting collection facility cost, and noise reduction facility cost as the costs of the port pollution prevention and control strategies corresponding to the port pollution prevention and control indicator, and calculate the water quality improvement revenue, reduction in waste treatment volume revenue, and increase in fish and bird population revenue as the benefits of the port pollution prevention and control strategies corresponding to the port pollution prevention and control indicator; calculate the operation cost of reusing dredged materials and / or construction waste, and rainwater reuse cost as the costs of the port resource reuse strategies corresponding to the port resource reuse indicator, and calculate the revenue from reusing dredged materials and rainwater reuse revenue as the benefits of the port resource reuse strategies corresponding to the port resource reuse indicator; calculate the life-cycle cost of the artificial fish release project and the life-cycle cost of the plant planting project as the costs of the eco-friendly strategies corresponding to the eco-friendly indicator, and calculate the fishery resource restoration revenue, coastal protection benefit, and ecosystem service restoration benefit as the benefits of the eco-friendly strategies corresponding to the eco-friendly indicator.

[0019] In an embodiment of the present invention, by way of example, taking "Port B" as an example, the server starts to execute the calculation of the life-cycle cost-benefit results of the corresponding strategies for each first-level port green correlation index. For the energy-saving and low-carbon strategies corresponding to the energy-saving and low-carbon indicators, the server calculates various costs and benefits. In terms of energy facility costs, it statistics the procurement, transportation and installation costs of port electric handling equipment, hybrid machinery, etc., as well as the construction, installation and supporting costs of energy supply facilities such as charging stations and hydrogen refueling stations. The annual energy consumption cost is calculated based on the energy consumption of electric, hydrogen energy and other equipment and the corresponding energy prices, combined with the annual operating time of the equipment. The annual operation and maintenance cost covers the consumption material costs such as equipment consumables and fuel, the equipment repair and regular maintenance costs, and the wages of operation and maintenance personnel. The equipment depreciation cost is allocated according to the service life of the equipment and energy supply facilities by the straight-line depreciation method or the double declining balance method. In terms of benefit calculation, by reducing traditional fuel consumption and shore power usage, the energy cost savings are calculated. Based on the reduction in emissions, the benefits brought by the reduction in air pollutants are calculated. For the port efficiency indicator strategies corresponding to the port efficiency indicators, the server calculates the multimodal internal transfer cost as the cost. This includes the infrastructure transformation costs of railway and port connection facilities, railway special lines and handling equipment, etc., as well as the cost of the optimized layout of the transfer loading and unloading sites. The benefits are calculated from three aspects: the increased throughput revenue, which is obtained from the increase in the annual throughput of the port after multimodal optimization and the unit throughput profit; the land use optimization revenue, which is calculated from the port land area released by the optimized land layout, the land market value and the percentage increase in utilization rate; the internal transfer efficiency improvement revenue, which is obtained based on the reduction in internal transportation time, the economic value of time to the cargo turnover and the total annual cargo handling volume of the port. In the port pollution prevention and control strategies corresponding to the port pollution prevention and control indicators, the server calculates the domestic sewage pipe network connection cost, including the costs of pipeline design, pipe materials, installation and pump station equipment purchase, as well as the subsequent operation and maintenance costs and depreciation costs. The cost of waste classification and collection facilities covers the construction costs of trash cans and waste classification stations, as well as the operating costs such as garbage truck fuel, manual operation and equipment maintenance. The cost of noise reduction facilities includes the construction costs of forest land or noise barriers, as well as the subsequent maintenance and repair costs and depreciation costs. In terms of benefits, the water quality improvement revenue is calculated through the amount of treated sewage, the reduction in pollutant concentration and the market value of unit pollutants. The waste treatment volume reduction revenue is calculated from the reduction in the amount of classified waste, the waste treatment cost and the value of recyclable waste. According to the changes in the affected fish and bird populations and related indicators, the fish and bird population increase revenue is calculated. For the port resource reuse strategies corresponding to the port resource reuse indicators, the server calculates the operation costs of reusing dredged materials and / or construction waste, including dredging, transportation, screening and treatment, reuse transformation and environmental impact mitigation costs. The rainwater reuse cost covers the collection system, storage equipment, treatment facilities, distribution system, maintenance system and energy consumption costs.In terms of benefit calculation, the benefits of reusing dredged materials include landfill cost savings, reduced new material procurement costs, policy incentives, and environmental benefits. The benefits of rainwater reuse include municipal water resource cost savings, reduced sewage treatment costs, policy subsidies, and ecosystem improvement benefits. For the eco-friendly strategies corresponding to the eco-friendly indicators, the server calculates the full life cycle costs of the fish stocking enhancement project, such as the costs of fry procurement, transportation, and manual stocking. The full life cycle costs of the plant planting project include the costs of seedling procurement and transportation, planting technical equipment, and subsequent maintenance and monitoring. In the benefit calculation, the benefits of fishery resource restoration are obtained based on the survival quantity, survival rate, fish weight, market price, and relevant coefficients of the stocked fry. The coastal protection benefits are calculated based on the mangrove area, wave reduction rate, flood and wind prevention value, and the lifespan of the mangroves. The benefits of ecosystem service restoration are obtained through the restored water area, enhanced biodiversity, improved water quality, and increased carbon sink and the unit ecological value.

[0020] In an embodiment of the present invention, the elastic net regression model is constructed based on the multiple first-level port green correlation indicators and the port greening degree, and the weight information corresponding to each of the first-level port green correlation indicators is calculated, which can be implemented through the following example.

[0021] Through the formula: , the weight information corresponding to each of the first-level port green correlation indicators is calculated; wherein, is the port greening degree; is the th first-level port green correlation indicator; is the regression coefficient, indicating the contribution of each first-level port green correlation indicator to the greening; and are respectively and regularized weight parameters.

[0022] In an embodiment of the present invention, by way of example, taking "Port C" as an example, the server undertakes the task of constructing an elastic net regression model based on multiple primary port green correlation indicators and the port greening degree, and calculating the weight information corresponding to each primary port green correlation indicator. The server needs to clarify the core idea of model construction: comprehensively consider the influence of each primary port green correlation indicator on the port greening degree, and at the same time use a special calculation method to avoid the model over-relying on certain indicators or inaccurate results caused by complex relationships between indicators. For "Port C", the primary port green correlation indicators include energy conservation and low-carbon indicators, port efficiency indicators, port pollution prevention and control indicators, port resource recycling indicators, and ecological friendliness indicators. The server regards these indicators as key factors affecting the port greening degree. During the calculation process, the server will perform a series of complex operations. It will analyze the correlation between each primary port green correlation indicator and the port greening degree. Just like when analyzing the energy conservation and low-carbon indicators, the server will consider how the combined influence of multiple secondary indicators such as the proportion of low-carbon energy and the proportion of terminals using renewable energy affects the port greening degree. To ensure the accuracy and stability of the model, the server adopts a method similar to setting rules for indicator weights. On the one hand, for situations where the model may be overly biased towards certain indicators, the server will make adjustments to avoid the model over-relying on certain indicators with too large weights, just as if preventing a certain indicator from being overly emphasized when evaluating the port greening. On the other hand, when there are complex correlations between some indicators that may lead to unstable calculation results, the server will also take measures to make the weight estimation of each indicator more stable and not show large fluctuations. After this series of operations and adjustments, the server finally obtains the weight information corresponding to each primary port green correlation indicator. For example, if the weight corresponding to the energy conservation and low-carbon indicator is relatively high, it means that in the greening process of "Port C", the performance in energy conservation and low-carbon has a greater impact on the improvement of the overall greening degree, and when formulating relevant strategies, it is necessary to focus on and prioritize resource investment in this aspect.

[0023] In an embodiment of the present invention, the constructing of the elastic net regression model based on the multiple primary port green correlation indicators and the port greening degree, and calculating the weight information corresponding to each of the primary port green correlation indicators can be implemented through the following example.

[0024] Through the formula: , the weight information corresponding to each of the primary port green correlation indicators is calculated; wherein, is divided into energy conservation and low-carbon indicators, port efficiency indicators, port pollution prevention and control indicators, port resource recycling indicators, and ecological friendliness indicators; is the secondary port green correlation indicator ; The contribution of the secondary port green correlation index to each primary port green correlation index; The contribution of each primary port green correlation index to the degree of greening, is the regularization hyperparameter.

[0025] In the embodiment of the present invention, by way of example, taking "Port D" as an example, the server starts to construct an elastic net regression model based on multiple primary port green correlation indexes and the port greening degree, and then calculates the weight information corresponding to each primary port green correlation index. The server first clarifies the key elements in the model. Among them correspond to the energy conservation and low-carbon index, port efficiency index, port pollution prevention index, port resource recycling index, and ecological friendliness index of Port D respectively. And each primary index is composed of multiple secondary port green correlation indexes constitute. For the energy conservation and low-carbon index , it includes 6 secondary port green correlation indexes such as the proportion of low-carbon energy and the proportion of terminals applying renewable energy ( to ). The server will analyze the contributions of these secondary indexes to the energy conservation and low-carbon index respectively, and this contribution degree is represented by . For example, the contribution of the secondary index of the proportion of low-carbon energy to the energy conservation and low-carbon index in the greening process is reflected by . Similarly, the port efficiency index is composed of 3 secondary port green correlation indexes such as the access rate of the inbound railway line ( to ), and the server analyzes the contributions of these secondary indexes to the port efficiency index . The port pollution prevention index , the port resource recycling index , and the ecological friendliness index also analyze the contributions of their respective secondary indexes in this way. Then, the server needs to consider the contributions of each primary port green correlation index ( to ) to the port greening degree, and this contribution is represented by . For example, the contribution of the energy conservation and low-carbon index to the overall greening degree of Port D is reflected by . In the whole calculation process, in order to make the model more accurate and stable, and avoid over-relying on certain indexes or resulting in result deviation due to the complex relationship between indexes, the server will use a method similar to an adjustment rule, which involves a regularization hyperparameter . The server performs a series of complex and meticulous calculations, comprehensively considering various secondary port green correlation indicators for the primary port green correlation indicators contribution , and the contribution of each primary port green correlation indicator to the port greening level . At the same time, it is adjusted in combination with the regularization hyperparameter . Finally, the server obtains the weight information corresponding to each primary port green correlation indicator, that is, the specific values of each . These values clearly show the respective importance degrees of the energy-saving and low-carbon indicators, port efficiency indicators, port pollution prevention and control indicators, port resource recycling indicators, and ecological friendliness indicators in the greening process of Port D. For example, if has a relatively large value, it indicates that the energy-saving and low-carbon indicators play a more crucial role in enhancing the greening level of Port D.

[0026] In the embodiment of the present invention, based on the full life cycle cost-benefit result and the weight information, the optimal solution of the green strategy combination for the target port is determined through a genetic algorithm, and the implementation can be carried out through the following examples.

[0027] Construct an objective function: ; wherein, is the total benefit of the port greening strategy composed of each strategy , is the total cost of the port greening strategy composed of each strategy , is the weight of each strategy determined based on elastic net regression, , , ; Based on the inverse function of the objective function, construct the fitness function of the genetic algorithm. Through the selection, crossover, and mutation operations of the genetic algorithm, when the maximum number of iterations is reached or the fitness value converges to a preset threshold, terminate the iteration and output the optimal solution of the green strategy combination for the target port.

[0028] In the embodiment of the present invention, by way of example, taking "Port E" as an example, the server starts to determine the optimal solution of the green strategy combination for this port through a genetic algorithm. First, the server constructs an objective function. This objective function needs to comprehensively consider the benefits, costs, and weights of each port greening strategy. For each greening strategy of Port E, the server counts all the benefits brought by this strategy, which is , such as the total benefits brought by energy cost savings and reduced air pollutants in the energy-saving and low-carbon strategy. At the same time, count all the costs during the implementation of this strategy, that is , such as the total of energy facility costs, annual energy consumption costs, etc. in the energy-saving and low-carbon strategy. And for each strategy 's weight is determined in a more complex way. It is obtained by multiplying and . is calculated based on the contribution of the secondary port green correlation index to the primary port green correlation index , specifically, it is the absolute value of divided by the sum of the absolute values of all . For example, under the energy-saving and low-carbon index, the corresponding to the secondary index of the proportion of low-carbon energy will participate in the calculation together with the corresponding to other secondary indexes to obtain . Then, it is calculated based on the contribution of each primary port green correlation index to the degree of greening , which is the absolute value of divided by the sum of the absolute values of all . For example, the corresponding to the energy-saving and low-carbon index is jointly calculated with the corresponding to other primary indexes to obtain . Finally, the weight of each strategy Through these calculations, the server constructs an objective function, whose purpose is to minimize a value that comprehensively considers benefits, costs, and weights. Then, the server constructs the fitness function of the genetic algorithm based on the inverse function of the objective function. This means that the higher the fitness, the lower the value of the objective function, indicating that higher greening benefits can be obtained at the same cost. Then, the server starts to perform operations of the genetic algorithm. In the selection step, the server selects better combinations from many possible combinations of green strategies as the parents according to the values of the fitness function. For example, among the many strategy combinations in Port E, those combinations that can improve the port greening benefits to a greater extent under a certain cost are more likely to be selected. After that is the crossover operation. The server crosses the selected parent strategy combinations, similar to the crossover combination of genes, to generate new offspring strategy combinations. For example, cross the combination focusing on energy-saving and low-carbon strategies with the combination focusing on port resource recycling strategies to form a new strategy combination method, expecting the new combination to have the advantages of both. At the same time, the server also performs mutation operations, making small random changes to some strategy combinations to avoid the algorithm falling into a local optimal solution. For example, slightly adjust the implementation degree of a certain strategy in a certain strategy combination. During this process, the server continuously performs iterations. When the preset maximum number of iterations is reached, or the fitness value converges to a preset threshold, the server considers that a relatively optimal solution has been found, then terminates the iteration, and outputs the optimal solution of the green strategy combination for Port E. This optimal solution can ensure that Port E achieves the maximum greening benefits through the synergistic effect of various greening strategies under limited resources, such as achieving economic sustainability while improving the environmental protection level.

[0029] In an embodiment of the present invention, the following implementation manners are further provided.

[0030] Optimize the objective function by using a time weight factor to obtain an optimized objective function: ; Wherein, is the time weight factor, , is the time weight adjustment coefficient, is the strategy 's implementation period, is the preset implementation period threshold; Construct the fitness function of the genetic algorithm based on the inverse function of the optimized objective function, and calculate and obtain the optimal solution of the green strategy combination of the target port.

[0031] In an embodiment of the present invention, by way of example, taking "Port F" as an example, during the process of the server determining the optimal solution of the green strategy combination, it will optimize the objective function by using the time weight factor. The server first clarifies the parameters related to the time weight factor. For each greening strategy of Port F , its implementation period needs to be determined . For example, in the energy-saving and low-carbon strategy, promoting new energy equipment, the entire period from equipment procurement, installation to commissioning is . Preset the implementation period threshold , which is a standard time value set according to the overall port planning and experience. The time weight adjustment coefficient is a value determined based on factors such as the port's preference for different implementation period strategies. Then, the server calculates the time weight factor for each strategy according to the formula . Suppose there is a resource recycling strategy in Port F, and its implementation period is short, less than the preset implementation period threshold . Then is negative, and under the action of the time weight adjustment coefficient , will be less than 1. This indicates that due to the short implementation period of this strategy, its weight in the objective function is relatively reduced. Conversely, if the implementation period of an eco-friendly strategy is long, greater than , , , will be greater than 1, and the weight of this strategy in the objective function is relatively increased. Then, the server optimizes the original objective function using the calculated time weight factor to obtain the optimized objective function. For example, for two strategies with similar benefits and costs, the strategy whose implementation period is more in line with the port planning expectations (with a small difference from ) will have a more reasonable comprehensive weight in the optimized objective function. After that, the server constructs the fitness function of the genetic algorithm based on the inverse function of the optimized objective function. Similar to before, the higher the fitness, the lower the value of the optimized objective function, indicating that after considering the time factor, higher greening benefits can be obtained at the same cost. Finally, the server performs iterative calculations according to the process of the genetic algorithm through selection, crossover, and mutation operations. In the selection step, the server will select the strategy combinations with high fitness as the parents. For example, those strategy combinations with better comprehensive benefits and costs after considering the time factor are more likely to be selected. During the crossover operation, the parent strategy combinations cross with each other to generate offspring, expecting the new combinations to combine the advantages of all parties. The mutation operation makes small random changes to some strategy combinations to prevent the algorithm from falling into a local optimum. When the maximum number of iterations is reached or the fitness value converges to the preset threshold, the server stops the iteration and outputs the optimal solution of the green strategy combination for Port F considering the time factor. This optimal solution enables Port F to not only balance benefits and costs in the greening process but also reasonably allocate resources according to the strategy implementation period to achieve more efficient green development.

[0032] To more clearly describe the solution provided by the embodiments of the present invention, the following provides a relatively complete implementation manner. Please refer to Figure 2 , Figure 2 , which is a schematic diagram of the framework of the intelligent optimization system for port green strategies based on elastic net regression weighting provided by the embodiments of the present invention.

[0033] The embodiments of the present invention construct a systematic port greening evaluation system to comprehensively evaluate the performance of ports in the process of greening. This system covers five dimensions: clean energy utilization, intensification ability, pollution prevention, resource recycling, and ecological friendliness, and is refined into 22 evaluation indicators, including the proportion of low-carbon energy, energy consumption intensity, coverage rate of sewage treatment facilities, noise compliance rate, and the number of ecological restoration projects. These indicators can guide the quantification of the greening progress of ports in multiple dimensions. To optimize the port greening level, the embodiments of the present invention specifically propose strategies for aspects such as clean energy utilization, intensification ability, pollution prevention, resource recycling, and ecological friendliness, and quantitatively evaluate their cost-benefits. To further improve the implementation effect of port greening strategies, the embodiments of the present invention introduce an optimization method based on the elastic net regression model. In this model, 22 specific evaluation indicators and 5 dimensions are used as independent variables, and the overall port greening level is used as the target variable. The weights of each strategy in the greening optimization model are obtained, which reflect the contribution of each strategy to the port greening level, and help decision-makers prioritize the strategies that have the most positive impact on the port greening progress. The embodiments of the present invention also introduce a genetic algorithm to preferentially retain the strategy combinations with better performance and continuously eliminate the relatively inferior combinations, and finally obtain the optimal solution that meets the port greening requirements.

[0034] (1)Construction of the port greening evaluation system; In the embodiments of the present invention, a systematic port greening evaluation system is constructed to comprehensively evaluate the multi-faceted performance of ports during the greening process. The constructed system is divided into multiple levels and covers five dimensions: clean energy utilization, port intensification ability, pollution prevention, resource recycling, and ecological friendliness. These 5 dimensions are further broken down into 22 specific evaluation indicators, including the proportion of low-carbon energy, energy consumption intensity, coverage rate of sewage treatment facilities, noise compliance rate, and the number of ecological restoration projects. The indicators are used to precisely quantify the greening progress of ports in specific areas and ensure a comprehensive reflection of the port greening level. (2)Construction of greening improvement strategies and their quantitative cost-benefit calculation formulas; In the embodiments of the present invention, specific improvement strategies are proposed for the port greening evaluation system, and the costs and benefits of relevant strategies are quantitatively analyzed. The greening improvement strategies mainly cover the following areas: In terms of clean energy and carbon emissions reduction, ports promote low-carbon energy such as electricity, hydrogen, and liquefied natural gas to replace traditional fossil fuels, thereby reducing carbon emissions and energy consumption. At the same time, supporting infrastructure such as charging facilities, hydrogen refueling stations, and LNG filling equipment is constructed. The costs of such strategies include multiple aspects such as equipment procurement, installation, maintenance, and depreciation, and the benefits are reflected in the savings in energy costs and the reduction of air pollutants. The resource recycling strategy involves the reuse of existing buildings and facilities in the port, including the use of recycled materials, the reuse of dredged soil, and the reuse of rainwater. The strategy costs are mainly concentrated on the construction, treatment, and maintenance of the resource recycling system, and the benefits are derived from reducing the procurement cost of new materials, reducing waste treatment costs, and the added value of environmental protection. In terms of pollution prevention strategies, ports effectively handle the pollutants generated during port operations by strengthening sewage treatment facilities, garbage classification collection systems, and hazardous waste isolation storage facilities. The benefits of such strategies include improving the environmental protection level of the port and achieving economic benefits by reducing pollutant emissions and lowering treatment costs. The ecological friendliness strategy focuses on protecting and restoring the ecological environment around the port through ecological structures (including breakwaters with fish reef functions) and ecological restoration projects. The benefits of the strategy include enhancing biodiversity and increasing the carbon sink capacity of the ecosystem. (3)Port greening strategy optimization model based on elastic net regression; To further optimize the implementation of port greening strategies, the embodiments of the present invention propose a green strategy optimization and selection method based on the weights of the elastic net regression model. The elastic net regression model can effectively handle the multicollinearity problem among various indicators in high-dimensional data and identify the key indicators that contribute the most to the greening goal through feature selection. In this model, 22 specific evaluation indicators and 5 dimensions are used as independent variables, and the overall greening level of the port is the target variable. The regression coefficients in the elastic net regression are normalized to obtain the optimization weights of the port greening strategy optimization model. In the intelligent optimization and selection system of green strategies, the independent variable is the benefit-cost ratio of port greening strategies, and the weights are calculated through the elastic net regression model.Through normalization, all weights are calculated to be between 0 and 1, and their sum is 1, serving as the weights for the target optimization problem. To ensure that the weights of different objectives in multi-objective optimization have a relatively consistent measurement standard, for... and The application of the genetic algorithm improvement strategy can enhance the efficiency of obtaining the optimal result of the strategy selection. The genetic algorithm simulates the process of natural selection and optimizes the combination of different greening strategies through multiple rounds of iteration to finally obtain the optimal solution. The optimal solution can not only ensure the synergistic effect of various strategies but also achieve the maximum greening benefit under the limited cost conditions.

[0035] Step 1: The evaluation index system for the port greening level (using this system to calculate weights and identify the prominent problems of a specific port in terms of greening).

[0036] In the embodiments of the present invention, an index system for the port greening degree is constructed. By defining five primary indicators and 22 secondary indicators, the port greening degree is quantitatively evaluated from multiple aspects, covering aspects such as energy conservation and low carbon, port efficiency, port pollution prevention and control, port resource recycling, and ecological friendliness. And a quantitative calculation formula is constructed for each indicator. In the constructed greening evaluation framework, the energy conservation and low-carbon index system mainly focuses on the use of low-carbon energy and the optimization of port energy consumption. The proportion of low-carbon energy is evaluated by statistically calculating the proportion of energy such as electricity and LNG in the total consumption; the application of renewable energy promotes the use of green energy by the proportion of renewable energy terminals, reduces carbon emissions, and realizes sustainable development. The energy consumption intensity index quantifies the energy consumption per unit throughput by comparing the total energy consumption of the port with the throughput, improving energy efficiency. The onshore power utilization rate reduces the energy consumption of the ship's self-provided power generation equipment by statistically calculating the proportion of ships using onshore power when berthing, thereby reducing air pollution and greenhouse gas emissions. The port also needs to increase the number of berths with onshore power supply to improve the facility coverage rate. In terms of port efficiency, indicators such as the access rate of inbound railway lines and the proportion of non-road collection and distribution are constructed in the port greening evaluation index system to measure the port logistics efficiency. By calculating the proportion of existing railway lines accessing the operation area, the popularity of railway transportation is evaluated, the pressure on road transportation is reduced, and an environmentally friendly and efficient logistics model is promoted. The key indicators for port pollution prevention and control include the coverage rate of municipal pipe networks and the closure rate of dry bulk cargo yards, etc., which evaluate the wastewater treatment capacity and the dust control effect respectively. The oil and gas recovery coverage rate and the compliance rate of port boundary noise reflect the port's performance in improving air quality and noise control. The solid waste classification collection rate and the coverage rate of ship pollutant receiving facilities measure the management level of the port in the treatment of solid waste and ship pollutants. The key indicators for port resource recycling include the cold energy utilization rate and the comprehensive utilization rate of dredged soil. The cold energy utilization rate is for LNG terminals to measure the efficient use of cold energy, and the comprehensive utilization rate of dredged soil reflects the reasonable reuse of dredged soil, promoting resource conservation and sustainable management. The ecological friendliness indicators are reflected in the number of ecological structure projects and the number of ecological restoration projects. The port aims to improve the port's ecological environment and enhance its ecological friendliness by implementing ecological structure and ecological restoration projects. Please refer to Table 1 for details. Table 1 is the index system and evaluation of the port greening degree.

[0037] Table 1

[0038] Step 2, the life-cycle cost-benefit assessment of the port greening strategy.

[0039] The embodiments of the present invention specifically propose improvement strategies for the port greening evaluation system and conduct quantitative analysis on the costs and benefits of relevant strategies. In terms of port clean energy and carbon emission reduction, the quantitative calculation of costs covers the expenses of equipment procurement, transportation, and installation, energy consumption, operation and maintenance costs, and equipment depreciation costs. The benefits are reflected in energy cost savings and reduced pollutant emissions. By reducing traditional fuel consumption, energy costs are lowered, and at the same time, environmental and economic benefits are brought about by reducing nitrogen oxide and sulfur dioxide emissions. In terms of resource recycling, the quantitative accounting of costs covers the reuse costs of existing buildings, structures, and recycled materials, as well as the screening, treatment, and reuse costs of dredging materials. Rainwater recycling involves the construction and maintenance costs of collection systems, storage facilities, and treatment facilities. In terms of benefits, it is mainly reflected in reducing the procurement costs of new materials, municipal water usage fees, and sewage treatment fees, while improving the resource recovery utilization rate and the ecological benefits of the port. In terms of pollution prevention and control, the quantitative accounting of costs covers the construction costs of connecting the domestic sewage pipe network to the municipal system or self-owned treatment facilities, the investment in garbage classification and collection facilities, and the construction and maintenance costs of noise reduction facilities. The benefits are reflected in improving the wastewater treatment efficiency, reducing the amount of landfill waste, lowering the environmental pollution risk, and protecting the ecological environment of fish and birds through noise reduction facilities, thereby enhancing the ecological benefits of the port. In terms of eco-friendly strategies, the quantitative accounting of costs covers the construction costs of structures with ecological functions (such as breakwaters, permeable structures), as well as the investment in artificial fish release and greening projects. The benefits are reflected in ecosystem restoration, increased biodiversity, and improved fish survival rate, bringing long-term economic returns to fishery resources. In terms of port efficiency improvement strategies, the cost calculation involves the construction of multimodal transport systems, land use optimization, and the transformation costs of internal transmission systems. The benefits are reflected in increasing the annual throughput, improving the land utilization rate, and accelerating the cargo turnover speed, thus bringing significant economic benefits to the port. Cost-benefit of strategies for clean energy and carbon emission reduction in the port. Port machinery and facilities use electricity or other clean energy, including hydrogen, liquefied natural gas (LNG), etc., and corresponding charging facilities or energy supply facilities are built accordingly; electric or hybrid loading and unloading and yard machinery / equipment / devices; auxiliary equipment / devices (including heating, cooling, and lighting) use >20% of clean energy (including solar water heaters, solar lighting, ground-source heat pumps, air-source heat pumps, heat exchangers, etc.). Costs: (1) Energy facility costs: The equipment procurement costs mainly include initial procurement and transportation and installation costs. The equipment specifically involves electric loading and unloading equipment, hybrid machinery, etc., as well as charging stations, hydrogen filling stations, LNG filling facilities, etc. (2) Annual energy consumption costs: Based on the energy consumption of the equipment used (electricity, hydrogen, or LNG, etc.), the annual consumption and energy prices of the comprehensive equipment are accounted for. (3) Annual operation and maintenance costs: The operation costs include material replacement, energy consumption, labor, maintenance, and repair, etc. The annual costs are adjusted according to actual maintenance needs.(4)Equipment depreciation cost: Depreciation is amortized annually according to the service life of equipment and energy supply facilities. The straight-line depreciation method or the double-declining balance method can be specifically adopted. Benefits: (1) Energy cost savings; (2) Reduction of air pollutants. In terms of port efficiency improvement, the cost-benefit of strategies includes seamless multimodal transportation, including rail-waterway, waterway-waterway, pipeline-waterway, and road-waterway. Multimodal transportation is considered in the water area and land layout. Optimize the land layout, build new connection lines, expand line capacity, etc. Establish an efficient internal transmission system (including appropriate layout and / or advanced dispatching system), etc. Costs: Port multimodal transportation costs include railways, port connection facilities, railway spurs, loading and unloading equipment, etc. for infrastructure transformation. And the cost of transfer loading sites for optimizing layout and improving efficiency. Benefits: (1) Benefits brought by increased throughput; (2) Benefits brought by optimized land use; (3) Benefits brought by improved internal transmission efficiency. In terms of resource recycling, the cost-benefit of strategies includes the reuse of existing buildings / facilities / structures; the use of recycled materials (including recycled steel, concrete, or wood); the reuse of dredged materials and / or construction waste, or transfer to professional companies; the reuse of treated sewage and rainwater for road spraying, bulk cargo yard spraying, greening, etc. Costs: (1) Reuse of dredged materials and / or construction waste; (2) Cost accounting for rainwater reuse. Benefits: (1) Benefits from the reuse of dredged materials; (2) Benefits from rainwater reuse. In terms of pollution prevention, the cost-benefit of strategies includes connecting domestic sewage pipelines to municipal drainage pipelines or self-owned treatment facilities; waste sorting and collection facilities; isolation facilities and sites for collecting, storing, and transporting hazardous waste; noise reduction facilities (including forest land, noise barriers), etc. Costs: (1) Connecting domestic sewage pipelines to municipal drainage pipelines or self-owned treatment facilities; (2) Waste sorting and collection facilities; (3) Noise reduction facilities. Benefits: (1) Benefits of connecting domestic sewage pipelines to municipal drainage pipelines or self-owned treatment facilities: Benefits of improved water quality. : Social and economic benefits of improving water quality by reducing sewage discharge. (2) Benefits of waste sorting and collection facilities: Benefits of reduced waste treatment volume : By sorting waste, the amount of waste finally entering landfills or incinerators is reduced, saving treatment costs. (3) Ecological benefits of noise reduction facilities for fish and birds: Benefits of fish survival : After noise reduction, the interference with fish behavior (including foraging, breeding, migration) is reduced, the survival rate of fish is increased, bringing ecological and economic benefits. Benefits of bird survival :After the noise is reduced, the living environment of birds will be improved, the interference of noise on birds’ foraging, reproduction, and migration behavior will be reduced, and the population will be increased. The cost-benefit of the eco-friendly strategy includes structures with ecological functions (including breakwaters and permeable structures with fish reef functions); budgets for artificial fish release and greening. Cost: (1) Cost accounting of the entire life cycle of the enhancement and release project; (2) Cost accounting of the entire life cycle of the plant planting project. Benefits: (1) Benefits of fishery resource restoration (R 1 ); (2) Coastal protection benefits (R 2 ); (3) Ecosystem service restoration benefits (R3).

[0040] Table 2 shows the cost-benefit evaluation system of port greening strategy Table 2

[0041] Step three: Identify the sensitivity of port greening indicators and intelligently optimize and select green solutions.

[0042] Sensitivity identification of greening indicators based on elastic net regression model: The elastic net regression model is adopted, and the degree of port greening is taken as the main target variable. The indicators are shown in Table 1. 22 specific indicators (including port clean energy utilization, port intensive capacity, port pollution treatment, resource recycling and eco-friendliness) are considered, and the contribution of each indicator is quantified through two-layer analysis. The formula organically combines 22 secondary indicators, 5 second-level factors and the degree of port greening Y, comprehensively considers the contribution of each indicator to port greening, and controls the complexity and fit of the model through regularization. The model can screen out unimportant indicators while ensuring the accuracy of the model, thereby simplifying the model and avoiding overfitting, so as to accurately identify the key indicators that contribute the most to port greening. In other words, due to the different characteristics of different ports, the prominent aspects of greening are also different, resulting in different contributions of various indicators to port greening. For sensitive indicators, the weight is greater when optimizing cost-effectiveness. By introducing two regularization methods, elastic net regression can not only perform effective feature selection when quantifying the contribution of multiple indicators to port greening, but also reduce the model's sensitivity to multicollinearity and avoid overfitting caused by excessive model complexity. Regularization has the effect of variable selection. By penalizing the absolute value of the regression coefficients, it avoids the model relying on some features with very large weights, thus preventing overfitting. When there is multicollinearity among features, regularization helps to stabilize the coefficient estimation of the model and avoid drastic fluctuations in the regression coefficients. The model formula: ; where: is the degree of port greening (target variable); is the th greening index (including clean energy utilization, etc.); is the regression coefficient, indicating the contribution of each index to greening; and are respectively and the weight parameters of regularization, controlling the model complexity and feature selection.

[0043] Integrating all levels, the complete elastic net regression model for the degree of port greening can be expressed as: where: (are the port capabilities in five aspects respectively). Each corresponds to an index and is the regression variable of five comprehensive factors, including clean energy utilization of the port , intensive port capacity , port pollution treatment , resource recycling and ecological friendliness . These indices are obtained by the weighted combination of their respective secondary specific indices. The elastic net regression is used to determine their respective contributions to the degree of greening. The secondary indices (22 specific indices) respectively represent the key parameters of all aspects of port greening and correspond to the 22 aspect indices for evaluating the degree of port greening in Table 1. The contribution of the index to each index ( to to ) is represented by the regression coefficient . Each is the regression coefficient of the index (such as clean energy utilization of the port and other five aspects), indicating the contribution of each second-layer factor to the degree of port greening .

[0044] Step 4, intelligent optimization and comparison of the benefits of green strategies based on the elastic net regression weights.

[0045] In the intelligent optimization and comparison selection system of green strategies, the independent variable is the benefit-cost ratio of port greening strategies (the indicators are shown in Table 2), and the weights are calculated using the elastic net regression model with the indicators in Table 1 in 3.2.3.1. Through normalization, all weights are calculated to be between 0 and 1, and their sum is 1, serving as the weights for the target optimization problem. To ensure that the weights of different objectives in multi-objective optimization have a relatively consistent measurement standard, for and . The genetic algorithm is applied to improve the efficiency of obtaining the optimal result of strategy comparison selection. The genetic algorithm simulates the process of natural selection and optimizes the combination of different greening strategies through multiple rounds of iteration to finally obtain the optimal solution. The optimal solution can not only ensure the synergistic effect of various strategies but also achieve the maximum greening benefit under the limited cost conditions. The objective function is: ; where: is the benefit brought by strategy , expressed as the sum of various benefits; is the cost of strategy , expressed as the sum of various costs; is the weight of each strategy determined from the elastic net regression. The weights are calculated using the elastic net regression model. Through normalization, all weights are calculated to be between 0 and 1, and their sum is 1, serving as the weights for the target optimization problem. To ensure that the multi-index weights have a relatively consistent measurement standard, the regression coefficients and in the elastic net regression are normalized. The weight normalization formula for secondary indicators is: ; The weight normalization formula for two-layer factors is: . The formula for calculating the weight of secondary indicators for the overall objective is: . Steps of the genetic algorithm: Population initialization: Randomly generate the initial solution space, using: , Fitness function: The fitness function uses the inverse function of the above objective function, that is, the higher the fitness, the lower the objective function value: Crossover and mutation: Select parents for crossover operation to generate offspring, and at the same time introduce a small amount of mutation to avoid local optimum. Termination condition: When the maximum number of iterations is reached or the fitness value converges to a certain threshold, the algorithm terminates and outputs the optimal solution.

[0046] Step Five, Recommendation of Port-Targeted Green Index Improvement and Optimization Strategies.

[0047] Based on the sensitivity identification of the aforementioned elastic net regression model and the intelligent optimization and comparison of greening strategies, it is recommended to prioritize the implementation of the strategies that contribute the most to port greening according to the differentiated weights of indicators. By combining the elastic net regression model and the genetic algorithm, the actual contribution of each greening strategy to port greening is quantified, and weight normalization is performed on each strategy, so that each strategy has a clear ranking and priority in the overall optimization. For different areas of port greening, including the utilization of clean energy, the improvement of intensification ability, pollution prevention and control, resource recycling, and ecological friendliness, the recommended optimization strategies will be intelligently compared according to their respective benefit-cost ratios. Specifically, it is recommended to prioritize the implementation of strategies that significantly improve the degree of greening and have a high cost-benefit ratio, including increasing the application proportion of clean energy, improving the energy efficiency management system, and optimizing pollutant emission control equipment. These strategies have a relatively high initial investment cost, but their long-term benefits can be significantly rewarded by reducing carbon emissions, improving energy efficiency, and reducing pollutant emissions. For strategies in areas such as pollution prevention and control and resource recycling, it is recommended to prioritize the construction of resource recycling systems, including the reuse of dredged soil and the optimization of rainwater recycling systems. These strategies can help the port achieve circular economy, improve the utilization rate of resources, and at the same time reduce the costs of waste treatment and new resource procurement. In addition, in terms of pollution prevention and control, the upgrading and maintenance of sewage treatment and air pollutant control systems should be strengthened to ensure the maximization of the port's environmental protection benefits. In terms of ecological friendliness, it is recommended to prioritize the implementation of infrastructure construction with ecological functions, including ecological breakwaters, permeable structures, etc., to enhance the port's ecological protection ability and promote the restoration of biodiversity. Such strategies can not only improve the greening level of the port, but also bring additional social and environmental benefits to the port. Based on the identification of indicator weights of the elastic net regression model and the coupling of the genetic algorithm optimization process, the combination of greening strategies is continuously optimized, and the final optimal solution can maximize the port greening goal under the premise of controlling costs. The optimized strategies will comprehensively improve the port's performance in the greening process, enabling it to make significant progress in the utilization of clean energy, pollution prevention and control, resource recycling, etc., and at the same time achieving sustainable long-term economic benefits.

[0048] Please refer to Figure 3 , Figure 3An intelligent optimization device 110 for port green strategies based on elastic net regression weighting provided by an embodiment of the present invention includes: an acquisition module 1101, configured to acquire a plurality of first-level port green correlation indicators corresponding to a target port, and the plurality of first-level port green correlation indicators include energy conservation and low-carbon indicators, port efficiency indicators, port pollution prevention and control indicators, port resource recycling indicators, and ecological friendliness indicators; a calculation module 1102, configured to calculate respective port greening strategies corresponding to each of the first-level port green correlation indicators, and determine the life-cycle cost-benefit results of each of the port greening strategies; an optimization module 1103, configured to construct an elastic net regression model based on the plurality of first-level port green correlation indicators and the degree of port greening, and calculate the weight information corresponding to each of the first-level port green correlation indicators; and based on the life-cycle cost-benefit results and the weight information, determine an optimal solution of the green strategy combination for the target port through a genetic algorithm.

[0049] It should be noted that the implementation principle of the foregoing intelligent optimization device 110 for port green strategies based on elastic net regression weighting can refer to the implementation principle of the foregoing intelligent optimization method for port green strategies based on elastic net regression weighting, which will not be elaborated here. An embodiment of the present invention provides a computer device 100, and the computer device 100 includes a processor and a non-volatile memory storing computer instructions. When the computer instructions are executed by the processor, the computer device 100 executes the foregoing intelligent optimization device 110 for port green strategies based on elastic net regression weighting. As Figure 4 shown, Figure 4 is a structural block diagram of the computer device 100 provided by an embodiment of the present invention. The computer device 100 includes an intelligent optimization device 110 for port green strategies based on elastic net regression weighting, a memory 111, a processor 112, and a communication unit 113. An embodiment of the present invention provides a readable storage medium, and the readable storage medium includes a computer program. When the computer program runs, it controls the computer device where the readable storage medium is located to execute the foregoing intelligent optimization device 110 for port green strategies based on elastic net regression weighting. For illustrative purposes, the foregoing description has been made with reference to specific embodiments. However, the above illustrative discussion is not intended to be exhaustive or to limit the disclosure to the precise form disclosed. According to the above teachings, numerous modifications and variations are possible. These embodiments are selected and described to best illustrate the principles of the disclosure and its practical applications, so that those skilled in the art can best utilize the disclosure and utilize various embodiments with different modifications to suit the specific applications expected.

Claims

1. An intelligent optimization method for port green strategy based on elastic net regression weighting, characterized in that: include: Acquire multiple first-level port green-related indicators corresponding to the target port, wherein the multiple first-level port green-related indicators include energy-saving and low-carbon indicators, port efficiency indicators, port pollution prevention and control indicators, port resource recycling indicators, and eco-friendly indicators; Calculate the port greening strategy corresponding to each of the first-level port green-related indicators respectively, and determine the full life cycle cost-effectiveness results of each of the port greening strategies; An elastic net regression model is constructed based on the plurality of first-level port green-related indicators and the degree of port greening, and weight information corresponding to each first-level port green-related indicator is calculated; Based on the full life cycle cost-effectiveness results and the weight information, an optimal solution for the green strategy combination for the target port is determined through a genetic algorithm.

2. The method according to claim 1, characterized in that The step of obtaining multiple first-level port green related indicators corresponding to the target port includes: The proportion of low-carbon energy, the proportion of terminals using renewable energy, energy consumption intensity, the proportion of berths available for shore power, shore power utilization rate and other energy supply capabilities are used as multiple secondary port green related indicators of the energy-saving and low-carbon indicators; The access rate of railway lines to the port, the proportion of non-road collection and distribution, and the length of time ships stay in the port are used as multiple secondary port green related indicators of the port performance indicators; The municipal pipe network coverage rate, dry bulk cargo yard closure rate, oil and gas recovery coverage rate, port boundary noise compliance rate, port solid waste classification collection and disposal rate, ship pollutant reception facility coverage rate and ballast water reception facility construction are used as multiple secondary port green related indicators of the port pollution prevention and control indicators; Taking cold energy utilization rate, dredged soil comprehensive utilization rate and unconventional water resource utilization as multiple secondary port green related indicators of the port resource reuse index; The number of projects that adopt ecological structures, the number of ecological restoration and rehabilitation projects, and special protection plans for important organisms will be used as multiple secondary port green related indicators of the above-mentioned eco-friendly indicators; Based on the multiple secondary port green association indicators, multiple primary port green association indicators corresponding to the target port are constructed.

3. The method according to claim 1, characterized in that The green port strategy corresponding to each of the first-level port green related indicators is calculated respectively, and the full life cycle cost-effectiveness result of each of the port green strategies is determined, including: Calculate the energy facility cost, annual energy consumption cost, annual operation and maintenance cost, and equipment depreciation cost as the cost of the energy-saving and low-carbon strategy corresponding to the energy-saving and low-carbon index, and calculate the energy cost savings and air pollutant reduction as the benefits of the energy-saving and low-carbon strategy corresponding to the energy-saving and low-carbon index; Calculating the intermodal internal transshipment cost as the cost of the port efficiency indicator strategy corresponding to the port efficiency indicator, and calculating the throughput improvement benefit, land use optimization benefit and internal transmission efficiency improvement benefit as the benefits of the port efficiency indicator strategy corresponding to the port efficiency indicator; Calculate the cost of domestic sewage pipe network connection, garbage classification collection facility cost and noise reduction facility cost as the cost of the port pollution prevention and control strategy corresponding to the port pollution prevention and control index, and calculate the benefits of water quality improvement, garbage disposal volume reduction and fish and bird population increase as the benefits of the port pollution prevention and control strategy corresponding to the port pollution prevention and control index; Calculate the operating cost of reusing dredged materials and / or construction waste and the cost of rainwater reuse as the cost of the port resource reuse strategy corresponding to the port resource reuse index, and calculate the income from reusing dredged materials and the income from rainwater reuse as the benefits of the port resource reuse strategy corresponding to the port resource reuse index; The full life cycle cost of the stocking and release project and the full life cycle cost of the plant planting project are calculated as the costs of the eco-friendly strategies corresponding to the eco-friendly indicators, and the fishery resource recovery benefits, coastal protection benefits and ecosystem service restoration benefits are calculated as the benefits of the eco-friendly strategies corresponding to the eco-friendly indicators.

4. The method according to claim 1, characterized in that The elastic net regression model is constructed based on the plurality of first-level port green-related indicators and the port greening degree, and the weight information corresponding to each first-level port green-related indicator is calculated, including: By formula: , calculate and obtain the weight information corresponding to each of the first-level port green related indicators; in, The greening degree of the port; For the First-level port green related indicators; is the regression coefficient, indicating the contribution of each first-level port green-related indicator to greening; and They are and The weight parameter of the regularization.

5. The method according to claim 2, characterized in that: The elastic net regression model is constructed based on the plurality of first-level port green-related indicators and the port greening degree, and the weight information corresponding to each first-level port green-related indicator is calculated, including: By formula: , calculate and obtain the weight information corresponding to each of the first-level port green related indicators; in, It is divided into energy-saving and low-carbon indicators, port efficiency indicators, port pollution prevention and control indicators, port resource recycling indicators and eco-friendly indicators; Green related indicators for secondary ports ; Green related indicators for secondary ports through their contribution to green related indicators of each primary port; The contribution of each first-level port green-related indicator to the greening degree. is the regularization hyperparameter.

6. The method according to claim 5, characterized in that The method of determining the optimal solution of the green strategy combination for the target port based on the full life cycle cost-benefit result and the weight information through a genetic algorithm includes: Construct the objective function: ; in, For each strategy The benefits constitute the sum of the benefits of the port greening strategy. For each strategy The benefits constitute the sum of the costs of the port greening strategy, For each strategy determined in elastic net regression The weight of , , ; The fitness function of the genetic algorithm is constructed based on the inverse function of the objective function. Through the selection, crossover and mutation operations of the genetic algorithm, when the maximum number of iterations is reached or the fitness value converges to a preset threshold, the iteration is terminated and the optimal solution of the green strategy combination of the target port is output.

7. The method according to claim 6, characterized in that The method further comprises: The objective function is optimized using the time weight factor to obtain the optimized objective function: ;in, is the time weight factor, , is the time weight adjustment coefficient, For strategy implementation cycle, To preset the implementation cycle threshold; The fitness function of the genetic algorithm is constructed based on the inverse function of the optimized objective function, and the optimal solution of the green strategy combination of the target port is calculated.

8. A port green strategy intelligent optimization device based on elastic net regression weighting, characterized in that: include: An acquisition module, used to acquire a plurality of first-level port green-related indicators corresponding to the target port, wherein the plurality of first-level port green-related indicators include energy-saving and low-carbon indicators, port efficiency indicators, port pollution prevention and control indicators, port resource recycling indicators, and eco-friendly indicators; A calculation module, used to calculate the port greening strategy corresponding to each of the first-level port green-related indicators, and determine the full life cycle cost-effectiveness results of each of the port greening strategies; The optimization module is used to construct an elastic net regression model based on the multiple first-level port green-related indicators and the port greening degree, and calculate the weight information corresponding to each of the first-level port green-related indicators; based on the full life cycle cost-effectiveness results and the weight information, determine the optimal solution of the green strategy combination for the target port through a genetic algorithm.

9. A computer device, characterized in that: The computer device comprises a processor and a non-volatile memory storing computer instructions. When the computer instructions are executed by the processor, the computer device executes the method according to any one of claims 1 to 7.

10. A readable storage medium, characterized in that: The readable storage medium includes a computer program, and when the computer program is executed, the computer device where the readable storage medium is located is controlled to execute the method according to any one of claims 1 to 7.

Images