Comprehensive energy system carbon collaborative scheduling method, device and electronic equipment

Abstract

The present application relates to the technical field of integrated energy system optimization scheduling, and particularly relates to an integrated energy system energy-carbon collaborative scheduling method, device and electronic equipment. The method comprises the following steps: based on multi-source energy operation data and carbon emission factors of the integrated energy system, a multi-objective energy-carbon collaborative optimization model is constructed; the multi-objective energy-carbon collaborative optimization model is solved by using an information optimization falcon algorithm to obtain a Pareto optimal solution set; and a scheduling scheme of the integrated energy system is generated according to the Pareto optimal solution set. The present application can solve the problem that the existing integrated energy system energy-carbon scheduling method has limitations in dealing with multi-objective, multi-coupling and multi-uncertainty operation scenarios. The integrated energy system can still maintain efficient and stable operation when facing uncertainty and multi-coupling constraints.

Description

Technical Field

[0001] This invention relates to the field of integrated energy system optimization and scheduling technology, and in particular to a method, apparatus and electronic equipment for integrated energy system energy and carbon coordinated scheduling. Background Technology

[0002] As the energy structure continues to evolve towards multi-energy complementarity and system synergy, Integrated Energy Systems (IES) are gradually becoming the core form of power system and energy system research and engineering practice. IES centers on electricity, integrating multiple energy forms such as heat, cold, and gas energy to achieve coordinated exchange, mutual support, and efficient utilization among multiple energy flows. Internally, it includes various energy conversion and coupling devices such as Combined Cooling, Heating and Power (CCHP) units, gas turbines, electric chillers, absorption chillers, boilers, energy storage devices, and thermal storage systems. Significant coupling relationships exist between these devices, such as electro-thermal coupling, electro-gas coupling, and cold-thermal coupling. These complex coupling relationships result in the operation of IES exhibiting high-dimensional, multi-constraint, and strongly nonlinear characteristics, making traditional single-energy system scheduling methods difficult to directly apply to such multi-energy systems.

[0003] Meanwhile, the optimal scheduling of energy systems is no longer solely focused on economic efficiency; it must also comprehensively consider carbon emission constraints to achieve synergistic optimization between economic costs and carbon emissions. Taking traditional energy equipment such as gas turbines and boilers as examples, their fuel consumption directly affects carbon emissions, and carbon emission factors are related to energy type, operating status, and external environmental conditions. While renewable energy sources such as wind power and photovoltaics have near-zero emissions, their output is highly random and uncertain, requiring system scheduling to balance economic efficiency with carbon emission control and efficient renewable energy utilization. The superposition of multiple objectives, multiple couplings, and multiple constraints makes the energy-carbon synergistic optimization of integrated energy systems a typical high-dimensional NP-hard problem.

[0004] On the other hand, the increasing proportion of renewable energy has led to greater uncertainty in the operation of integrated energy systems. For example, photovoltaic output is significantly affected by irradiance and weather conditions, wind power is limited by wind speed fluctuations, and seasonal differences and real-time fluctuations in user loads significantly increase the complexity of system scheduling. In this context, scheduling optimization not only needs to ensure the system's economic efficiency and low carbon footprint but also maintain the reliability of its energy supply, ensuring that user-side demands for electricity, heating, cooling, and gas are met. Energy supply reliability, a long-standing key indicator, has a more complex composition in integrated energy system scenarios. Factors such as power redundancy, heating network capacity, gas pipeline stability, and energy storage support capabilities all affect system reliability assessment.

[0005] Faced with the aforementioned complexity, current research has gradually shifted from traditional mathematical programming models to the application of intelligent optimization algorithms. In existing research, algorithms such as Particle Swarm Optimization (PSO), Genetic Algorithm (GA), Ant Colony Optimization (ACO), Simulated Annealing (SA), and Grey Wolf Optimizer (GWO) are widely used in energy system scheduling problems. However, these algorithms generally suffer from limited search capabilities, over-reliance on the initial population, susceptibility to local optima, or insufficient convergence speed. In practical engineering scenarios, due to the large scale of integrated energy systems, the numerous devices, and the complex constraints, traditional intelligent algorithms often struggle to obtain high-quality scheduling solutions within a reasonable timeframe, especially when dealing with multi-objective optimization and multi-period coupled constraints, where their performance disadvantages are even more pronounced. Summary of the Invention

[0006] This invention provides a method, apparatus, and electronic device for coordinated energy and carbon scheduling of integrated energy systems, addressing the limitations of existing integrated energy system energy and carbon scheduling methods when handling multi-objective, multi-coupling, and multi-uncertainty operating scenarios. This invention enables integrated energy systems to maintain efficient and stable operation even when facing uncertainties and multi-coupling constraints.

[0007] This invention provides a method for coordinated energy and carbon scheduling in an integrated energy system, comprising: Based on the multi-source energy operation data and carbon emission factors of the integrated energy system, a multi-objective energy-carbon synergistic optimization model is constructed. The multi-objective energy-carbon co-optimization model was solved using the information-optimized Falcon algorithm to obtain the Pareto optimal solution set; Based on the Pareto optimal solution set, a scheduling scheme for the integrated energy system is generated.

[0008] In one embodiment, the construction of a multi-objective energy-carbon synergistic optimization model based on multi-source energy operation data and carbon emission factors of an integrated energy system includes: The multi-source energy operation data and carbon emission factors of the integrated energy system are used as inputs; Define an objective function, which includes at least the following objectives: minimizing economic costs, minimizing carbon emissions, and maximizing energy supply reliability; wherein, economic costs include electricity purchase costs, fuel costs, and equipment operation and maintenance costs; carbon emissions are calculated based on the carbon emission factors and consumption of various energy sources; and energy supply reliability is measured based on the ratio of the load deficit of electrical energy, heat energy, and cooling energy to the total load. The constraints are set, including at least energy balance constraints, coupling constraints, equipment capacity constraints, and ramp constraints. Energy balance constraints include the real-time supply and demand balance of electrical, thermal, and cooling energy. Coupling constraints include the ratio of electrical and thermal output of combined cooling, heating, and power (CCHP) units, and the ratio of thermal and cooling output of absorption chillers. Equipment capacity constraints include the upper and lower limits of output or energy storage for each energy device. Ramp constraints include restrictions on the rate of change of output of each energy device in adjacent time periods. By combining the objective function and the constraints, the multi-objective energy-carbon collaborative optimization model is obtained.

[0009] In one embodiment, before solving the multi-objective energy-carbon synergistic optimization model using the Information Optimization Falcon algorithm, the method further includes: The population size, maximum number of iterations, inertia factor, and information interaction parameters of the information optimization falcon algorithm are set, and an initial population is generated based on the decision variable range of the multi-objective energy-carbon collaborative optimization model. Calculate the information entropy of the initial population. If the information entropy is lower than a preset threshold, perform replacement or perturbation operations on some individuals to obtain an enhanced population. An information exchange network is established in the enhanced population. During the algorithm iteration process, the information exchange network is used to provide each individual with the optimal location information of its neighboring individuals. The neighboring individuals refer to other individuals that are directly connected to the individual through the information exchange network.

[0010] In one embodiment, each individual in the Information Optimized Falcon Algorithm corresponds to a set of candidate scheduling schemes for the multi-objective energy-carbon co-optimization model; the step of solving the multi-objective energy-carbon co-optimization model using the Information Optimized Falcon Algorithm to obtain the Pareto optimal solution set includes: In the k-th iteration, the position of the individual after information exchange is determined based on the current position of the individual, the global optimal position of the individual in the current iteration, and the optimal position of the neighboring individuals obtained through the information exchange network. The position of the individual after the information exchange, the worst individual position in the current iteration, and a random number based on the Levy distribution are used to determine the position of the individual after the jump. Based on the individual position after the jump, the elite individual position in the current iteration, and a random number following a standard normal distribution, the individual position after local development is determined, where the elite individual position is the average position of the most fit individuals in the current population. Based on the convergence trend of the current iteration process, the inertia factor and information interaction parameters of the information-optimized falcon algorithm are adaptively adjusted to obtain the updated algorithm parameters. Using the individual position after local development as the initial position of the individual in the next iteration, and using the updated algorithm parameters as the initial parameters for the next iteration, perform the (k+1)th iteration until the iteration termination condition is met, and output the Pareto optimal solution set.

[0011] In one embodiment, generating a scheduling scheme for the integrated energy system based on the Pareto optimal solution set includes: Using the analytic hierarchy process, we determined the corresponding weights for economic cost, carbon emission, and energy supply reliability for the three indicators: economic cost, carbon emission, and energy supply reliability. Based on the economic cost weight, carbon emission weight, and energy supply reliability weight, calculate the comprehensive evaluation value of each solution in the Pareto optimal solution set; The solution with the highest comprehensive evaluation value is selected from the Pareto optimal solution set as the final scheduling scheme.

[0012] In one embodiment, the method further includes: The generated scheduling scheme is then distributed to the energy equipment of the integrated energy system for execution. Collect actual operating data after scheduling execution, and use the actual operating data as input data for constructing the multi-objective energy-carbon synergistic optimization model in the next round of optimization.

[0013] The present invention also provides an integrated energy system carbon coordination scheduling device, comprising the following modules: The modeling module is used to construct a multi-objective energy-carbon synergistic optimization model based on multi-source energy operation data and carbon emission factors of integrated energy systems. The optimization module is used to solve the multi-objective energy-carbon co-optimization model using the information optimization falcon algorithm to obtain the Pareto optimal solution set; The scheduling module is used to generate a scheduling scheme for the integrated energy system based on the Pareto optimal solution set.

[0014] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the integrated energy system energy-carbon coordinated scheduling method as described above.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the integrated energy system energy-carbon coordinated scheduling method as described above.

[0016] The present invention also provides a computer program product, including a computer program which, when executed by a processor, implements the integrated energy system energy-carbon collaborative scheduling method as described in any one of the above.

[0017] The integrated energy system energy-carbon collaborative scheduling method, device and electronic device provided by the present invention construct a multi-objective energy-carbon collaborative optimization model based on the multi-source energy operation data and carbon emission factors of the integrated energy system, comprehensively considering multiple objectives such as economic cost, carbon emission and system energy supply reliability, and can better adapt to multi-objective operation scenarios and meet the scheduling requirements of the integrated energy system under complex operating conditions. The information optimization falcon algorithm is used to solve the multi-objective energy-carbon collaborative optimization model. After obtaining the Pareto optimal solution set, a scheduling plan for the integrated energy system can be generated according to this solution set. This enables the integrated energy system to maintain efficient and stable operation according to the generated scheduling plan when facing uncertainty and multi-coupling constraints, solves the limitations of the prior art in dealing with multi-uncertainty operation scenarios, and improves the stability and reliability of system operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0019] Figure 1 is a schematic flowchart of the integrated energy system energy-carbon collaborative scheduling method provided by the present invention.

[0020] Figure 2 is a schematic diagram of the overall architecture of the integrated energy system energy-carbon collaborative scheduling device provided by the present invention.

[0021] Figure 3 is a schematic overall flowchart of the information optimization falcon algorithm provided by the present invention.

[0022] Figure 4 is a schematic diagram of the energy flow coupling relationship in the integrated energy system provided by the present invention.

[0023] Figure 5 is a schematic flowchart of instruction issuance provided by the present invention.

[0024] Figure 6 is a schematic diagram of the structure of the integrated energy system energy-carbon collaborative scheduling device provided by the present invention.

[0025] Figure 7 is a schematic diagram of the structure of the electronic device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0027] In recent years, a class of swarm intelligence algorithms inspired by animal behavior has attracted attention. Among them, falcon hunting behavior, characterized by leaping exploration, encirclement and approach, and rapid ambush, is considered suitable for high-dimensional and complex search spaces. However, the original falcon algorithm still suffers from shortcomings such as low efficiency in utilizing population information, lack of parameter adaptability, and oscillating convergence, making it difficult to directly apply to the scheduling of complex multi-objective energy systems. Especially in the context of energy-carbon synergy, the scheduling problem not only requires solving a high-dimensional search space but also dynamic trade-offs among different objectives, making its complexity far exceed that of general optimization problems.

[0028] Currently, some research has attempted to introduce information theory or entropy theory to improve intelligent algorithms, such as enhancing the diversity of the initial population through information entropy and adjusting the search direction through information gain, thereby enabling the algorithm to have higher global exploration capabilities and stronger robustness. However, most research remains at the theoretical level and lacks a systematic approach that combines energy and carbon coordinated scheduling of integrated energy systems. In addition, there is currently a lack of an overall optimization framework that can simultaneously integrate information interaction mechanisms, adaptive dynamic adjustment strategies, and swarm intelligence behavior, specifically designed to solve the multi-objective energy and carbon scheduling needs of integrated energy systems.

[0029] Based on this, under the trend of intelligent and low-carbon development of energy systems, there is an urgent need for an intelligent optimization algorithm that can fully utilize multi-source information, possess strong global search capabilities and rapid convergence capabilities, and adapt to complex multi-objective optimization environments, along with the construction of a supporting integrated energy and carbon optimization device. This device not only needs to complete data acquisition and model building, but also needs to achieve comprehensive coordination of economy, low carbon emissions, and energy supply reliability through intelligent algorithms, enabling the integrated energy system to maintain efficient and stable operation even in the face of uncertainties and multiple coupling constraints. In view of this, the present invention provides an integrated energy system energy and carbon coordinated scheduling method, device, and electronic equipment.

[0030] This invention aims to address the limitations of existing integrated energy system carbon scheduling methods in handling multi-objective, multi-coupling, and multi-uncertainty operating scenarios. It proposes a multi-objective collaborative optimization scheme that simultaneously considers economic efficiency, carbon emissions, and energy supply reliability, and constructs an integrated energy system carbon collaborative optimization device based on this scheme. This device achieves joint optimization scheduling of multiple energy systems (electricity, heat, cooling, and gas) through real-time acquisition of multi-source energy operation data, construction of a multi-objective carbon collaborative optimization model, and solution using an intelligent optimization algorithm based on an information interaction enhancement mechanism. The core of this invention lies in introducing an information-optimized falcon algorithm. By embedding an information-theoretic driving mechanism, a population structure optimization mechanism, and a dynamic feedback adjustment structure into the traditional swarm intelligence optimization strategy, the algorithm significantly enhances its global exploration and local development capabilities in high-dimensional, multi-coupling search spaces. This enables it to effectively address the contradictions and complex constraints between different objectives in integrated energy system scheduling problems, improving the stability, robustness, and accuracy of the optimization solution.

[0031] The device described in this invention first monitors in real time the operating status of the power system (such as electrical load, generator output, and predicted renewable energy power), thermal system parameters (such as heating load, heat network return water temperature, and waste heat recovery device efficiency), refrigeration system operating conditions, and gas source pressure and flow information of the gas system involved in the integrated energy system, and simultaneously acquires the carbon emission factors of different energy carriers. Based on this, the collected data is cleaned, missing data is filled in, anomalies are removed, and time series alignment is performed to ensure the accuracy and consistency of the data required for subsequent optimization models. Because there are coupling relationships between the various energy flows in the integrated energy system and strong inter-device linkages, data preprocessing is not only a data quality control measure but also a fundamental condition for model construction, providing reliable input data for multi-objective optimization solutions.

[0032] After data preparation, the device of this invention constructs a multi-objective energy-carbon collaborative optimization model for the integrated energy system through a modeling module. This model comprehensively considers objectives such as economic cost, carbon emissions, and energy supply reliability. Economic cost typically includes power generation fuel costs, heat source costs, cold source costs, electricity purchase costs, and energy storage system loss costs; carbon emissions are calculated based on the carbon emission factors and output power of different energy devices; energy supply reliability is related to reserve capacity, heating capacity, gas supply capacity, and the ability of energy storage to participate in regulation. The model also includes numerous operational constraints, including equipment power constraints, electricity-heat-cooling-gas coupling constraints, energy storage state constraints, thermal balance constraints, gas grid pressure constraints, power grid flow constraints, and multi-period scheduling constraints. This optimization problem exhibits typical nonlinear, multivariable, and multi-constraint characteristics, and there may be significant conflicts between different objectives. For example, minimizing economic cost often leads to increased carbon emissions, while prioritizing renewable energy consumption may increase system reserve demand, thereby affecting energy supply reliability. Therefore, a smart optimization algorithm with strong global performance is required for solving the problem.

[0033] To solve the aforementioned multi-objective energy-carbon collaborative optimization model, this invention constructs an optimization module based on the information-optimized Falcon algorithm. Compared to the traditional Falcon optimization algorithm, this invention introduces an information entropy-driven mechanism in the initialization phase. By calculating the distribution differences of the population in the search space, it increases the diversity of the initial population, enabling the algorithm to cover a wider range of solution spaces in the early stages of the search. During the population evolution process, this invention establishes an information exchange network among individual falcons, allowing different individuals to share their captured local optimal information, thereby achieving collaborative advancement of the group and improving global convergence efficiency. Building upon the traditional Falcon algorithm's strategies such as raids, encirclements, and jumps, this invention further incorporates a dynamic feedback adjustment mechanism based on convergence trends. Based on the algorithm's convergence speed, search stability, and objective function fluctuations during the current iteration, it dynamically adjusts parameters such as step size, jump coefficient, and inertia factor, allowing the algorithm to flexibly switch between exploration and development modes at different search stages. Furthermore, this invention combines multi-objective ranking and congestion assessment mechanisms to construct a Pareto optimal solution set, enabling the optimization results to form multiple operational schemes for schedulers to choose from, considering economic efficiency, carbon emissions, and energy supply reliability.

[0034] Once the optimization module finds the Pareto optimal solution that satisfies the constraints and reaches the set convergence condition, the scheduling module of the device of this invention will select one of the optimal solutions according to the actual operating requirements. It will then convert the start-up and shutdown plans, power output, heating and cooling capacity, energy storage charging and discharging arrangements, and renewable energy consumption strategies of each energy device within the system into executable instructions, which will be sent to the relevant equipment for execution through the energy management platform or field controller. This scheduling scheme ensures controllable economic costs and compliance with carbon emission policies, while also guaranteeing energy supply reliability, enabling the integrated energy system to maintain high efficiency and low carbon emissions under different operating conditions.

[0035] In summary, by embedding the information-optimized Falcon algorithm into the integrated energy system scheduling optimization framework, this invention not only effectively enhances the algorithm's global search capability and improves the solution quality, but also achieves a unified coordination of economy, low carbon emissions, and reliability through a multi-objective collaborative solution mechanism, thereby constructing an intelligent and low-carbon scheduling optimization device applicable to actual integrated energy systems.

[0036] The following is combined with Figures 1 to 7 This invention describes a method, apparatus, and electronic device for integrated energy system carbon-coordinated scheduling. By collecting and modeling the operating status, predicted load, and carbon emission data of multi-energy systems such as electricity, heat, cooling, and gas, this invention solves multi-objective energy-carbon scheduling schemes through an improved intelligent optimization algorithm. It can be widely applied to regional energy internet, integrated energy stations, multi-energy complementary systems, and industrial park energy management platforms.

[0037] Figure 1 This is a flowchart illustrating the integrated energy system carbon-coordinated scheduling method provided by the present invention, as shown below. Figure 1 As shown, the method includes the following: S110. Based on the multi-source energy operation data and carbon emission factors of the integrated energy system, a multi-objective energy-carbon synergistic optimization model is constructed.

[0038] Reference Figure 2As shown, based on sensors and communication modules deployed in the energy pipeline network, equipment controllers, and metering terminals, multi-source energy operation data is collected, including real-time electrical and thermal power and fuel consumption of combined cooling, heating, and power (CCHP) units; photovoltaic power plant output; wind power output; cooling power and energy consumption of electric chillers and absorption chillers; thermal power and gas consumption of gas boilers; and the state of charge and charging / discharging heat output of energy storage and thermal storage systems. Simultaneously, 24-hour rolling forecasts and real-time monitoring values ​​of the park's production and office electrical load, thermal load, and cooling load, as well as carbon emission factor data, are collected. The collected multi-source energy operation data, load data, and equipment status data undergo data preprocessing. The 3σ criterion is used to remove outlier data, wavelet transform is used to filter high-frequency noise in the load and renewable energy output data, and interpolation methods based on long short-term memory networks are used to fill in missing data to ensure data integrity. Subsequently, time-series alignment and dataset construction are performed, converting all data to a fixed time scale, such as 15 minutes, to construct a unified energy-carbon dataset containing equipment operating parameters, load data, carbon emission factors, and equipment status, generating input data for M time segments.

[0039] Based on this unified carbon dataset, a multi-objective collaborative optimization model is established, encompassing system operating economic costs, carbon emissions, and renewable energy utilization rates. Objective functions may include optimization targets such as energy procurement costs, equipment operation and maintenance costs, carbon emission costs, and energy utilization efficiency. The carbon emission target is calculated based on real-time and predicted multi-source energy operation data and corresponding carbon emission factors to determine the system's total carbon emissions. Other objectives are set according to actual operational needs. Constraints are set, including real-time balance constraints for electrical, thermal, and cooling power; upper and lower limits of output, ramp-up rates, and minimum start-up and shutdown times for each energy device; state of charge, charging and discharging heat release power, and capacity constraints for the energy storage system; and safe operation constraints for the power grid and heating network.

[0040] S120. The information optimization falcon algorithm is used to solve the multi-objective energy-carbon co-optimization model to obtain the Pareto optimal solution set.

[0041] The information-optimized Falcon algorithm is used to solve the constructed multi-objective energy-carbon synergistic optimization model. Based on the traditional Falcon algorithm, this algorithm introduces a population diversity maintenance mechanism based on information entropy, uses Pareto dominance relations for individual ranking, and adaptively adjusts the exploration and development balance parameters to efficiently search the high-dimensional solution space containing continuous and discrete variables. Ultimately, it obtains the Pareto optimal solution set representing the trade-offs between multiple objectives such as economy, low carbon emissions, and energy efficiency.

[0042] S130. Generate a scheduling scheme for the integrated energy system based on the Pareto optimal solution set.

[0043] The Pareto optimal solution set consists of multiple non-dominated solutions, each representing a complete system operation plan. These plans exhibit different trade-offs in emphasizing objectives such as reducing economic costs, reducing carbon emissions, and improving energy supply reliability. Based on the actual constraints of system operation and scheduling preferences, a decision rule is established to select the final plan from the Pareto optimal solution set. This decision-making process includes quantitatively evaluating and comparing the performance of each plan in the solution set, with evaluation dimensions covering indicators such as economic cost, system operation plan, and energy supply reliability. Through comprehensive comparison or multi-attribute decision-making methods, the solution that best meets the current scheduling requirements is determined as the final scheduling plan.

[0044] The decision variable values ​​included in the final selected scheduling scheme are converted into device-level control commands that can be executed by the integrated energy system. This conversion process maps the abstract optimization results into specific time-series operation commands, including the planned output values, start-stop status settings, and interactive power plans between the system and the external power grid or gas grid at each time segment within the scheduling cycle, thereby generating a final scheduling scheme that can be directly issued to the integrated energy system for execution.

[0045] Existing energy system optimization and scheduling technologies focus solely on economic efficiency, neglecting carbon emission constraints. Furthermore, with increasing renewable energy consumption and heightened system uncertainty, it struggles to balance economic efficiency, low carbon emissions, and reliable energy supply. This invention constructs a multi-objective energy-carbon co-optimization model based on multi-source energy operation data and carbon emission factors from an integrated energy system. This model comprehensively considers multiple objectives, including economic cost, carbon emissions, and system energy supply reliability, better adapting to multi-objective operating scenarios and meeting the scheduling needs of integrated energy systems under complex operating conditions. The information optimization Falcon algorithm is used to solve the multi-objective energy-carbon co-optimization model. After obtaining the Pareto optimal solution set, a scheduling scheme for the integrated energy system can be generated based on this solution set. This enables the integrated energy system to maintain efficient and stable operation under uncertainties and multiple coupling constraints, overcoming the limitations of existing technologies in handling multi-uncertainty operating scenarios and improving the stability and reliability of system operation.

[0046] In one embodiment, the construction of a multi-objective energy-carbon synergistic optimization model based on multi-source energy operation data and carbon emission factors of an integrated energy system includes: S1101, The multi-source energy operation data and carbon emission factors of the integrated energy system are used as inputs; S1102. Define an objective function, which includes at least the following objectives: minimizing economic costs, minimizing carbon emissions, and maximizing energy supply reliability; wherein, economic costs include electricity purchase costs, fuel costs, and equipment operation and maintenance costs; carbon emissions are calculated based on the carbon emission factors and consumption of various energy sources; and energy supply reliability is measured based on the ratio of the load deficit of electrical energy, thermal energy, and cooling energy to the total load. S1103. Set constraints, which include at least energy balance constraints, coupling constraints, equipment capacity constraints, and ramp constraints; wherein, energy balance constraints include real-time supply and demand balance of electrical energy, thermal energy, and cold energy; coupling constraints include the ratio of electric and thermal output of combined cooling, heating, and power units, and the ratio of thermal and cold output of absorption chillers; equipment capacity constraints include the upper and lower limits of output or energy storage of each energy device; ramp constraints include the limitation on the rate of change of output of each energy device in adjacent time periods; S1104. Combining the objective function and the constraints, the multi-objective energy-carbon collaborative optimization model is obtained.

[0047] Specifically, when constructing a multi-objective energy-carbon synergistic optimization model based on multi-source energy operation data and carbon emission factors of an integrated energy system, the collected multi-source energy operation data and carbon emission factors of the integrated energy system are first used as model inputs. This data includes, but is not limited to, real-time electrical / thermal power of CCHP units, fuel gas consumption, photovoltaic power plant and wind power output (including predicted and real-time values), cooling power and energy consumption of electric chillers / absorption chillers, thermal power and gas consumption of gas boilers, SOC status and charging / discharging / heating power of energy storage / thermal storage systems, 24-hour rolling predicted and real-time monitoring values ​​of industrial park production / office electrical / thermal / cooling loads, carbon emission factors of various energy sources, and equipment status data.

[0048] Subsequently, an objective function is defined, comprising three sub-objectives: minimizing economic costs, minimizing carbon emissions, and maximizing energy supply reliability. The economic cost objective function comprehensively considers electricity purchase costs, fuel costs, and equipment operation and maintenance costs, achieving economic optimization by minimizing the total cost. The carbon emission objective function is calculated based on the carbon emission factors and consumption of various energy sources, aiming to reduce carbon emissions during system operation. The energy supply reliability objective function assesses the reliability of the system's energy supply by measuring the ratio of the load deficit of electricity, heat, and cooling to the total load, ensuring that energy demands on the user side are met. Simultaneously, constraints are set to ensure the feasibility and effectiveness of the model solution. These constraints include energy balance constraints, ensuring real-time supply and demand balance of electricity, heat, and cooling within the system; coupling constraints, defining the proportional relationship of electrical and thermal output of combined cooling, heating, and power (CCHP) units, reflecting the physical coupling characteristics between equipment; equipment capacity constraints, specifying the upper and lower limits of output or energy storage for each energy device to prevent equipment overload or operation in inefficient zones; and ramp-up constraints, limiting the output variation of each energy device between adjacent scheduling periods to ensure the stability and technical feasibility of equipment operation. For example, the power change rate of combined cooling, heating and power units and gas-fired boilers is ≤10% of rated power / 15 minutes, and the power change rate of energy storage charging and discharging is ≤5MW / 15 minutes. Based on the above objective functions and constraints, a multi-objective energy-carbon synergistic optimization model is constructed.

[0049] The multi-objective energy-carbon synergistic optimization model of this invention achieves synergistic optimization of integrated energy systems among economic costs, carbon emissions, and energy supply reliability. By comprehensively considering operational data and carbon emission factors from multiple energy forms, the model can more accurately reflect the actual operating status of the system, providing a reliable basis for optimized scheduling. The definition of the objective function and the setting of constraints enable the model to balance the conflicts between different objectives during the solution process, finding a globally optimal or near-globally optimal solution.

[0050] In one embodiment, before solving the multi-objective energy-carbon synergistic optimization model using the information-optimized Falcon algorithm described in S120 above, the method further includes: The population size, maximum number of iterations, inertia factor, and information interaction parameters of the information optimization falcon algorithm are set, and an initial population is generated based on the decision variable range of the multi-objective energy-carbon collaborative optimization model. Calculate the information entropy of the initial population. If the information entropy is lower than a preset threshold, perform replacement or perturbation operations on some individuals to obtain an enhanced population. An information exchange network is established in the enhanced population. During the algorithm iteration process, the information exchange network is used to provide each individual with the optimal location information of its neighboring individuals. The neighboring individuals refer to other individuals that are directly connected to the individual through the information exchange network.

[0051] Specifically, before using the Information Optimization Falcon Algorithm to solve the multi-objective energy-carbon co-optimization model, it is necessary to set the algorithm parameters and generate the initial population. First, based on the problem complexity and computational resources, the population size, maximum number of iterations, inertia factor, and information interaction parameters of the Information Optimization Falcon Algorithm are set. The population size determines the algorithm's search range, the maximum number of iterations limits the algorithm's running time, the inertia factor affects the algorithm's global and local search capabilities, and the information interaction parameters control the degree of information sharing among individuals within the population. Subsequently, based on the range of decision variables in the multi-objective energy-carbon co-optimization model, an initial position is randomly generated for each falcon. These decision variables include, but are not limited to, CCHP unit output, gas boiler output, start-up and output of electric chillers / absorption chillers, energy storage / thermal storage charging and discharging power, and purchased / grid-connected power.

[0052] After generating the initial population, its information entropy is calculated to assess the population's diversity. If the information entropy is lower than a preset threshold, it indicates that the individuals in the population are too concentrated and lack diversity. In this case, replacement or perturbation operations need to be performed on some individuals. Replacement refers to randomly generating new individuals to replace those with low information entropy, while perturbation involves making minor adjustments to some decision variables of existing individuals to increase population diversity. After these operations, an enhanced population is obtained, with its diversity significantly improved. In the enhanced population, an information exchange network is further established. An undirected graph structure is used, with each falcon connected to 3-5 neighboring individuals. Neighboring individuals are determined by ranking based on positional similarity, for example, selecting several individuals closest to the current individual's position. This network is used during algorithm iteration to provide each individual with the optimal position information of its neighboring individuals. Neighboring individuals refer to other individuals directly connected to the current individual through the information exchange network; information sharing among them helps the algorithm converge to the global optimum more quickly during the search process.

[0053] This invention provides a solid foundation for subsequent multi-objective optimization by setting the core parameters of the information-optimized Falcon algorithm and generating an initial population. Parameter setting ensures a balance between global search and local exploitation, improving solution efficiency. The diversity enhancement operation of the initial population effectively avoids the algorithm getting trapped in local optima, increasing the probability of finding the global optimum. The establishment of an information exchange network further promotes information sharing among individuals within the population, enabling the algorithm to converge to the vicinity of the optimum more quickly during iteration. These preprocessing steps significantly improve the performance and stability of the information-optimized Falcon algorithm when solving multi-objective energy-carbon co-optimization models.

[0054] Compared to traditional intelligent algorithms, the information-optimized Falcon algorithm of this invention has significant advantages. Compared to the shortcomings of particle swarm optimization (PSO) which is prone to getting trapped in local optima and genetic algorithms which have slow convergence speeds, this algorithm, by introducing an information exchange mechanism and the aforementioned diversity enhancement operations, achieves stronger global search capabilities, faster convergence speeds, and better optimization stability. This algorithm can obtain a higher-quality Pareto optimal solution set with a wider coverage and more uniform distribution within a reasonable time, thus providing a better solution foundation for subsequent multi-attribute decision-making. This improvement significantly enhances both the solution quality and computational efficiency of the entire optimization decision-making process. The algorithm can more effectively explore the solution space and find optimal solutions that meet multiple objectives such as economic cost, carbon emissions, and energy supply reliability. Simultaneously, because the algorithm considers population diversity and information sharing during the solution process, its robustness is also significantly improved, maintaining stable solution performance under different initial conditions and operating environments.

[0055] In one embodiment, each individual in the Information Optimization Falcon Algorithm corresponds to a set of candidate scheduling schemes in the multi-objective energy-carbon co-optimization model, covering parameters such as energy equipment start-up and shutdown status, output allocation, and energy flow direction. The algorithm iteratively optimizes the individual positions, gradually approaching the optimal solution set that satisfies multiple objective constraints such as economic cost, carbon emissions, and energy supply reliability. The process of solving the multi-objective energy-carbon co-optimization model using the Information Optimization Falcon Algorithm to obtain the Pareto optimal solution set includes: S1201. In the k-th iteration, the position of the individual after information exchange is determined based on the current individual's position, the global optimal individual position of the current iteration, and the optimal position of the neighboring individuals obtained through the information exchange network. S1202. Determine the position of the individual after the information exchange, the worst individual position in the current iteration, and the random number based on the Levy distribution. S1203. Based on the individual position after the jump, the elite individual position of the current iteration, and the random number following a standard normal distribution, determine the individual position after local development. The elite individual position is the average position of the individuals with the best fitness in the current population. S1204. Based on the convergence trend of the current iteration process, the inertia factor and information interaction parameters of the information-optimized falcon algorithm are adaptively adjusted to obtain the updated algorithm parameters. S1205. Using the individual position after local development as the initial position of the individual in the next iteration, and using the updated algorithm parameters as the initial parameters for the next iteration, perform the (k+1)th iteration until the iteration termination condition is met, and output the Pareto optimal solution set.

[0056] The algorithm iteration process consists of four steps: "information exchange and update - rapid response jump - local development - dynamic feedback adjustment", as detailed below (taking the k-th iteration as an example): Information exchange and update: Each falcon updates its position by combining its current position, the best position of its neighbors (i.e., the position with the highest fitness among its neighbors), and the current global best position (the position of the individual with the highest fitness in all iterations). The weights are controlled by an inertia factor and an information interaction parameter; the former determines the degree to which the current position is retained, while the latter determines the intensity of absorbing neighborhood and global information.

[0057] Rapid-response jump: Simulating the ambush behavior of a falcon hunting, the algorithm adjusts its position by jumping after information exchange. The jump direction is based on the difference between the current position and the worst individual position (the individual position with the lowest fitness in all iterations), and the step size is generated by Levy distribution random numbers to enhance the algorithm's ability to escape local optima.

[0058] Local development: A refined search is performed on the position after the jump. The positions of the 10% of individuals with the best fitness in the current population are averaged to obtain the elite individual's position. For each individual in the current population, its new position after local development is determined by the individual's position after exchanging information through the preceding jump, the elite individual's position, and a random number following a standard normal distribution.

[0059] Dynamic feedback adjustment: The convergence trend is assessed by calculating the rate of change of fitness at the globally optimal individual position over the last 10 iterations. If the rate of change is below a first threshold (e.g., 5%), it indicates that the convergence speed is too slow; if it is above a second threshold (e.g., 20%), it indicates convergence oscillation. The inertia factor and information interaction parameters are dynamically adjusted based on the convergence trend. If convergence is too slow, the inertia factor is reduced to decrease dependence on the current position, while the information interaction parameters are increased to enhance neighborhood information sharing; if convergence oscillates, the parameters are adjusted in the opposite direction to stabilize the search process.

[0060] Repeat steps S1201-S1205 until k reaches the maximum number of iterations (e.g., 150) or the convergence accuracy meets the requirements, then terminate the iteration. From the previous iterations, select non-dominated solutions (i.e., solutions for which no other solution is superior in all objectives) to form the Pareto optimal solution set. Each solution corresponds to a set of scheduling schemes that satisfy multi-objective constraints, allowing decision-makers to choose according to actual needs.

[0061] Through the above implementation methods, the Information Optimization Falcon Algorithm, when solving multi-objective energy-carbon co-optimization models, can fully utilize information sharing among individuals within the population to achieve a balance between global search and local exploitation. Specifically, by combining the current individual position, the globally optimal individual position, and the optimal position of neighboring individuals for position updates, the algorithm can gradually approach the global optimum while maintaining population diversity. Utilizing Levy-distributed random numbers for jumps enhances the algorithm's ability to escape local optima and improves global search efficiency. Through local exploitation guided by elite individual positions, the algorithm can perform refined searches near the optimal solution, improving solution accuracy. Simultaneously, adaptively adjusting algorithm parameters based on convergence trends allows the algorithm to flexibly switch between exploration and exploitation modes at different search stages, further enhancing solution efficiency and stability. The final Pareto optimal solution set provides multiple feasible scheduling schemes for multi-objective energy-carbon co-optimization of integrated energy systems, meeting diverse needs in practical engineering.

[0062] In one embodiment, the step of generating a scheduling scheme for the integrated energy system based on the Pareto optimal solution set in S130 above includes: S1301. Using the analytic hierarchy process, determine the corresponding weights for economic cost, carbon emission, and energy supply reliability for the three indicators: economic cost, carbon emission, and energy supply reliability. S1302. Calculate the comprehensive evaluation value of each solution in the Pareto optimal solution set based on the economic cost weight, carbon emission weight, and energy supply reliability weight. S1303. Select the solution with the highest comprehensive evaluation value from the Pareto optimal solution set as the final scheduling scheme.

[0063] First, a hierarchical model is constructed, with the evaluation of integrated energy system dispatch schemes as the target layer and economic cost, carbon emissions, and energy supply reliability as the criteria layer. Through expert scoring or historical data statistics, pairwise comparison judgment matrices are constructed: for economic cost and carbon emissions, if the importance of economic cost is three times that of carbon emissions, the corresponding position in the matrix is ​​assigned a value of 3; similarly, comparisons between other indicators are performed. The eigenvalue method is used to calculate the largest eigenvalue and corresponding eigenvector of the judgment matrix. After normalizing the eigenvectors, the weights of each indicator are obtained; for example, the weight of economic cost is 0.5, the weight of carbon emissions is 0.3, and the weight of energy supply reliability is 0.2, and the sum of the weights is 1.

[0064] For each solution in the Pareto optimal solution set, its economic cost, carbon emissions, and energy supply reliability are extracted. Based on the determined weights, a weighted summation method is used to calculate the comprehensive evaluation value: multiply the economic cost of a solution by its weight of 0.5, the carbon emissions by its weight of 0.3, and the energy supply reliability by its weight of 0.2, and then add these three values ​​together to obtain the comprehensive evaluation value of the solution. For example, if a solution has an economic cost of 1 million yuan, carbon emissions of 50 tons, and an energy supply reliability of 98%, its comprehensive evaluation value is 100 × 0.5 + 50 × 0.3 + 98 × 0.2 = 90.6.

[0065] The algorithm iterates through the Pareto optimal solution set, comparing the comprehensive evaluation values ​​of all solutions to determine the solution with the highest value. For example, if solution A has a comprehensive evaluation value of 92.5, solution B has 90.6, and solution C has 88.3, ​​then solution A is selected as the final scheduling scheme. This scheme needs to specify the operating parameters of each energy device, such as gas turbine output, energy storage device charging and discharging power, and renewable energy consumption ratio, to ensure that multi-objective constraints are met.

[0066] This invention quantifies the relative importance of indicators using the analytic hierarchy process (AHP), transforming qualitative analysis into quantitative calculation and avoiding the arbitrariness of subjective weighting. This method accurately reflects the differences in decision-makers' preferences for economic, environmental, and reliability objectives under different scenarios, ensuring that weight allocation highly aligns with actual needs and providing an objective and interpretable evaluation basis for scheduling schemes. The comprehensive evaluation value calculation integrates economic cost, carbon emissions, and energy supply reliability indicators, ensuring that the final scheme achieves a dynamic balance among multiple objectives. Compared to single-objective optimization, the scheduling scheme generated by this method can simultaneously meet the requirements of cost control, environmental compliance, and stable operation, significantly improving the overall operational efficiency of the integrated energy system.

[0067] In one embodiment, the method further includes: The generated scheduling scheme is then distributed to the energy equipment of the integrated energy system for execution. Collect actual operating data after scheduling execution, and use the actual operating data as input data for constructing the multi-objective energy-carbon synergistic optimization model in the next round of optimization.

[0068] Specifically, the finalized scheduling plan is transformed into equipment control commands. This includes: parsing the operating parameters of each energy device in the scheduling plan, such as the output power of the gas turbine, the charging and discharging mode of the energy storage device, and the proportion of renewable energy power generation; encapsulating the control commands into standard data frames according to the equipment communication protocol; sending the commands to the controllers of the corresponding devices through wired or wireless communication networks; and after receiving the commands, the equipment controllers adjust their operating status to the target parameters, such as adjusting valve opening, changing motor speed, or switching circuit connection methods, to ensure that the actual operation is consistent with the scheduling plan.

[0069] During the execution of the scheduling scheme, real-time equipment operation data is collected through a sensor network, including but not limited to: the input / output power of energy equipment (such as the power consumption and heat generation of electric boilers), the state of charge changes of energy storage devices, the real-time power generation of renewable energy (such as the output voltage and current of photovoltaic arrays), and environmental parameters (such as outdoor temperature and light intensity). The collected data is synchronously integrated according to timestamps to form a real-world operation dataset containing multi-dimensional information. In the next round of optimization, this dataset is used as input to update the constraints (such as equipment efficiency curves and energy price fluctuations) and objective function parameters (such as dynamic adjustment of carbon emission factors) of the multi-objective energy-carbon co-optimization model, ensuring that the model can reflect the actual operating characteristics of the system and improve the optimization accuracy of subsequent scheduling schemes.

[0070] This invention enables a multi-objective energy-carbon co-optimization model to capture real-time dynamic changes in the system through a feedback mechanism based on actual operational data. These changes include efficiency declines due to equipment aging, seasonal load fluctuations, and energy market price adjustments. By continuously updating model parameters, the scheduling scheme generated in the next round of optimization is more closely aligned with actual operating conditions, reducing optimization errors caused by static model assumptions and forming a closed-loop improvement path of "optimization-execution-feedback-re-optimization." Model iteration based on operational data can gradually accumulate system operating patterns. Once this empirical data is incorporated into the model, it can guide subsequent scheduling schemes to more accurately balance economic costs and carbon emissions, achieving a long-term cumulative improvement in the operational efficiency of the integrated energy system and enhancing the system's adaptability and robustness to complex operating conditions.

[0071] Example 1: This embodiment uses a comprehensive energy system in a medium-sized manufacturing industrial park as the application object. The park includes two combined cooling, heating, and power (CCHP) units (gas turbine driven, rated electric power 10MW / unit, rated thermal power 8MW / unit), a photovoltaic power station (installed capacity 5MW), a wind power micro-source (installed capacity 2MW), three electric chillers (each with a rated cooling power of 4MW), two absorption chillers (each with a rated cooling power of 3MW, driven by CCHP waste heat), one gas boiler (rated thermal power 12MW), an energy storage system (capacity 10MWh, charge / discharge power 5MW), and a thermal storage tank (capacity 8MWh, charge / discharge power 4MW). It mainly provides electricity, heat, cooling, and gas energy services to 12 production plants and 3 office buildings within the park. The typical daily maximum electrical load is 18MW, maximum thermal load is 10MW, and maximum cooling load is 12MW. Gas demand is mainly used for fuel supply to the CCHP units and the gas boiler. The following refers to... Figure 3 , Figure 4 and Figure 5 The process will be explained.

[0072] 1.1 The following methods were used to complete the multi-source data acquisition and preprocessing: Step S1.1.1, Data Acquisition: The following data is collected through sensors and communication modules deployed in the park's energy pipeline network, equipment controllers, and metering terminals: Energy operation data: real-time electrical / thermal power and fuel consumption of CCHP units, photovoltaic power plant output (15-minute interval forecast + real-time value), wind power output (15-minute interval forecast + real-time value), cooling power and energy consumption of electric chillers / absorption chillers, thermal power and gas consumption of gas boilers, and SOC (State of Charge) status and charging / discharging / heat dissipation power of energy storage / thermal storage systems; Load data: 24-hour rolling forecasts and real-time monitoring values ​​of the park's production / office power load, heat load, and cooling load; Carbon emission factor: The carbon emission factor for natural gas is taken as 2.08 kg CO2 / m³. 3 The carbon emission factor for electricity purchased from the power grid is 0.58 kg CO2 / kWh (the annual average emission factor of the local power grid), and the carbon emission factor for photovoltaic / wind power is 0.03 kg CO2 / kWh (only considering the equipment operation and maintenance phase). Equipment status data: start / stop status of each energy device, fault signals, efficiency curves (such as the electrical efficiency curve of CCHP unit as a function of load rate, and COP value of chiller).

[0073] Step S1.1.2, Data Preprocessing: Abnormal data is removed using the "3σ criterion", such as abnormal values ​​where the output of the photovoltaic power station suddenly drops to 0 and there is no fault signal. High-frequency noise in the load and renewable energy output data is filtered by wavelet transform. For missing data (such as missing photovoltaic forecast data for a certain period of time), an interpolation method based on Long Short-Term Memory (LSTM) network is used to fill in the missing data to ensure data integrity. Step S1.1.3, Time Series Alignment and Dataset Construction: Convert all data to a 15-minute time scale (consistent with the park's scheduling cycle), and construct a unified energy and carbon dataset containing "equipment operating parameters - load data - carbon emission factors - equipment status". A total of 96 time segments of input data are generated for subsequent modeling and optimization.

[0074] 1.2 Based on the above data, a multi-objective energy-carbon synergistic optimization model is constructed, as follows: 1.2.1 The objective function includes: Economic Costs Minimize objective: Where t is the time segment (15 minutes / segment). The price of natural gas during time period t (taken as 3.2 yuan / m³). 3 ), The total gas consumption (m³) of the CCHP unit and gas boiler during time period t. 3 ), The grid purchase price for electricity during time period t is 0.85 yuan / kWh during peak hours, 0.52 yuan / kWh during normal hours, and 0.28 yuan / kWh during off-peak hours. The power purchased during time period t (kW). The price for surplus electricity fed into the grid during time period t is 0.38 yuan / kWh. The power supplied to the grid during time period t (kW) The equipment operation and maintenance cost for time period t is calculated at 0.02 yuan / (kW·h) based on the rated power of the equipment.

[0075] carbon emissions Minimize objective: in, The carbon emission factor for natural gas is 2.08 kg CO2 / m³. 3 ), Carbon emission factor for purchasing electricity from the grid (0.58 kg CO2 / kWh). The carbon emission factor for renewable energy is 0.03 kg CO2 / kWh. The total output of photovoltaic and wind power (kW) during time period t.

[0076] Power supply reliability Maximize the goal: in, , , These represent the electricity, heat, and cooling load deficits (kW / kWth / kWc) for time period t. , , These represent the total electrical, heating, and cooling loads (kW / kWth / kWc) for time period t, respectively. A coefficient of 0.01 is used to normalize the reliability index to the [0,1] interval (the closer the target value is to 1, the higher the reliability).

[0077] 1.2.2 Constraints include: Energy balance constraints: During time period t, the electrical load satisfies "CCHP power generation + photovoltaic + wind power + electricity purchase + electricity storage discharge = electrical load + electricity storage charging + surplus electricity grid connection", the heat load satisfies "CCHP heat generation + gas boiler heat generation + heat storage heat release = heat load + heat storage charging", and the cooling load satisfies "electric refrigeration + absorption refrigeration = cooling load". Equipment capacity constraints: CCHP unit electric power ≤ 10MW, thermal power ≤ 8MW, single electric chiller cooling power ≤ 4MW, single absorption chiller cooling power ≤ 3MW, energy storage SOC ∈ [20%, 80%], thermal storage SOC ∈ [15%, 85%]; Coupling constraints: The power-to-heat ratio of the CCHP unit is fixed at 1.25 (i.e., 1kW of power generation corresponds to 0.8kWth of heat production), and the heat-to-cooling ratio of the absorption chiller is fixed at 0.7 (i.e., 1kWth of heat consumption corresponds to 0.7kWc of cooling). Ramp-up constraints: The power change rate of CCHP units and gas-fired boilers is ≤10% of rated power / 15 minutes, and the power change rate of energy storage charging and discharging is ≤5MW / 15 minutes.

[0078] 1.3. Reference Figure 3 As shown, perform information optimization and Falcon algorithm initialization: Step S1.3.1, Parameter Settings: Falcon population size N=50, this setting balances search diversity and computational efficiency; maximum number of iterations K=150, testing shows convergence can be achieved in 150 iterations; inertia factor... The initial value is 0.9, and it decreases linearly to 0.4 with each iteration; information interaction parameters This parameter controls the intensity of information sharing among individuals; the local development ratio is 0.3, meaning that 30% of the population individuals focus on local search and 70% focus on global exploration. Step S1.3.2, Initial Population Generation: The algorithm's decision variables include "CCHP unit output, gas boiler output, electric chiller / absorption chiller start-up and output, energy storage / thermal storage charging and discharging power, and purchased / grid-connected power," totaling 22 decision variables. Based on the constraints of each variable (e.g., CCHP electric power ∈ [2MW, 10MW]), an initial position vector is randomly generated for each falcon. Discrete variables (such as equipment start-up and shutdown) are encoded in binary, while continuous variables (such as output) are encoded in real numbers; i is the individual index, representing the i-th candidate scheduling scheme; subscripts 1-22 represent the numbers of decision variables, each number corresponding to a specific control variable in the model (such as equipment start-up and shutdown, output, etc.).

[0079] Step S1.3.3, Enhancing Population Diversity: Calculate the information entropy of the initial population. ,in, Let E be the probability density of the position of the i-th falcon. If E < 1.2, the entropy value is too low, indicating insufficient population diversity. Then, the population distribution is optimized by "randomly replacing 10% of the individuals and perturbing the positions of 5% of the individuals". At the same time, an undirected information exchange network is constructed, where each falcon establishes a connection with 3-5 neighboring individuals (sorted by position similarity) for subsequent information sharing.

[0080] 1.4 Information optimization using the Falcon algorithm: The algorithm's iterative process consists of four steps: "information exchange and update - rapid response jump - local development - dynamic feedback adjustment," as detailed below (taking the k-th iteration as an example): Step S1.4.1, Falcon Information Exchange Update: Each falcon obtains the optimal position of its neighbors through the information exchange network. Combined with the globally optimal individual position Update its own location: ; Where k is the iteration number index. This represents the position of the i-th individual in the k-th iteration; The position of the individual after information exchange is represented by rand, which is a random number in the range [0,1]. In this embodiment, during the 30th iteration, Falcon 12 obtains the local optimal position of Falcon 8 in its neighborhood through the network, which is the optimal position of the aforementioned neighborhood individual. (Corresponding to "CCHP output 8MW, electricity purchase 3MW, energy storage and discharge 2MW"), combined with the globally optimal individual position (corresponding to "CCHP output 9MW, full photovoltaic absorption, heat storage and heat release 1MW"), the self-purchased power is adjusted from 4.5MW to 3.2MW, which is closer to the globally optimal direction.

[0081] Step S1.4.2, Falcon's Rapid Response Jump: Simulating the ambush behavior of a falcon during hunting, the individual positions are updated by jumping after information exchange, enhancing global search capabilities. ; In this context, the second number in the superscripts (k,1) and (k,2) is a stage identifier used to distinguish different update stages within a single iteration. The individual's position after the jump. =0.8 is the jump coefficient, and Levy(1.5) is a random number that follows a Levy distribution (exponent 1.5) and is used to generate non-Gaussian jump step size; The worst individual position in the k-th iteration; in this embodiment, in the 60th iteration, for the scenario of "photovoltaic output suddenly increases by 2MW", the Falcon 25 uses Levy jump to increase the energy storage charging power from 1.8MW to 3.5MW, quickly adapting to the fluctuation of renewable energy output and avoiding waste of surplus power.

[0082] Step S1.4.3, Local Development Mechanism: Perform a local fine-grained search on the individual's position after the jump to improve solution accuracy. ; in, For individual locations after partial development, Here, represents the local search coefficients, and randn is a random number following a standard normal distribution. The position of the elite individual in the k-th iteration can be the average position of the top 10% of the best individuals. In this embodiment, during the 90th iteration, for the period of "peak cooling load of 12MW", the elite individual position corresponds to "all 3 electric chillers are running (total cooling power of 12MW)". The Falcon 38 adjusts the output of electric chiller No. 1 from 3.8MW to 4.0MW and No. 2 from 4.2MW to 4.0MW through local development, so as to achieve precise matching of cooling load and reduce energy consumption.

[0083] Step S1.4.4, Dynamic Feedback Adjustment: Calculate the convergence trend evaluation function. (F(·) is the weighted value of the multi-objective function). This represents the current global optimal position, which is the position of the individual with the best fitness in the entire falcon population at the k-th iteration. The historical global optimal position is represented by the position at the k-th position. The position of the individual with the best fitness in the entire population after 10 iterations. A value <0.05 indicates that the convergence rate is too slow, so the inertia factor should be adjusted. Reduce information interaction parameters by 0.05. Increase by 0.03; if If the value is greater than 0.2, it indicates convergent oscillation; otherwise, adjustment is made. In this embodiment, during the 120th iteration, =0.042 (convergence slowed down), the algorithm automatically... From 0.5 to 0.45 The value increased from 0.6 to 0.63, further optimizing the global optimal individual position.

[0084] Repeat the above steps until the maximum number of iterations K=150, and finally obtain the Pareto optimal solution set containing 87 non-dominated solutions. The solution set covers a variety of trade-off schemes such as "low cost and high carbon emissions", "low carbon emissions and medium cost", and "high reliability and medium low carbon".

[0085] 1.5 Scheduling Scheme Generation and Execution: Step S1.5.1, Optimal Solution Selection: The Analytic Hierarchy Process (AHP) is used to evaluate the Pareto optimal solution set. Economic cost is weighted at 0.4, carbon emissions at 0.35, and energy supply reliability at 0.25. The comprehensive score of each solution is calculated, and the solution with the highest score is selected as the final scheduling scheme. This scheme has a total economic cost of 128,600 yuan, a total carbon emission of 85.2 tCO2, and an energy supply reliability of 99.82% within a typical 24-hour day. Step S1.5.2, Instruction Conversion: Convert the scheduling scheme into executable instructions for the equipment, for example: 00:00-06:00 (valley time), CCHP unit output 4MW (thermal power 3.2MW), gas boiler shut down, energy storage system charging (power 5MW, SOC increases from 20% to 80%), 2MW of electricity purchased; 08:00-12:00 (peak time), CCHP unit full output 10MW (thermal power 8MW), photovoltaic output 4.5MW, energy storage discharge 5MW, surplus electricity fed into the grid 1.2MW, 2 electric chillers operating (total cooling power 8MW). Step S1.5.3, Command Issuance: Commands are issued to the controllers of various devices, such as the CCHP unit PLC controller, the energy storage system EMS, and the power gateway metering device, via industrial Ethernet. The issuance delay is ≤1 second, and the command execution accuracy is 100%. Step S1.5.4: Feedback Iteration: Collect the actual operating data after execution (such as the actual output of CCHP unit 9.8MW, actual energy storage SOC 82%, and cooling load deficit 0MW). The deviation rate with the scheduling plan is ≤2%. Feed the actual operating data back to the data acquisition module as input data for the next round (next day) of optimization to form a closed-loop optimization.

[0086] 1.6 Implementation and Effect Verification: The method of this invention is compared with traditional particle swarm optimization (PSO) and genetic algorithm (GA). The results are as follows: In terms of economic costs, the method of this invention reduces costs by 6.8% compared to PSO and by 9.2% compared to GA, mainly due to the accurate selection of electricity purchase periods by the Information Optimization Falcon Algorithm (purchasing more electricity during off-peak hours and connecting to the grid more during peak hours) and the optimal matching of equipment output. In terms of carbon emissions, the method of this invention reduces emissions by 8.3% compared to PSO and 11.5% compared to GA. This is because the algorithm prioritizes the absorption of photovoltaic / wind power (the absorption rate is increased from 89% for PSO and 85% for GA to 96%), reducing the number of start-ups and shutdowns of gas-fired equipment. In terms of power supply reliability, the method of this invention improves by 0.3 percentage points compared with PSO and by 0.5 percentage points compared with GA, thanks to the rapid response of the local development mechanism to load fluctuations; In terms of convergence speed, the method of this invention has a convergence iteration count of 120 times, which is 25% less than PSO (160 times) and 33% less than GA (180 times). The dynamic feedback adjustment mechanism effectively avoids premature convergence of the algorithm.

[0087] Example 2: Co-optimization of energy and carbon in regional energy internet systems.

[0088] This embodiment uses a regional energy internet system in a new urban area as an example. The system covers an area of ​​5km². 2 It includes 2 regional power supply stations (each equipped with 1 CCHP unit, 2 electric chillers, and 1 gas boiler), distributed photovoltaic (total installed capacity 10MW), distributed wind power (total installed capacity 3MW), centralized energy storage (energy storage capacity 20MWh), regional heating network (supply and return water temperatures set at 95 / 50℃), and regional cooling network (supply and return water temperatures set at 7 / 12℃). The service targets include residential communities (5 in total), commercial complexes (2 in total), and science and technology parks (1 in total). The typical daily maximum electrical load is 35MW, the maximum heat load is 22MW, and the maximum cooling load is 28MW. The grid access voltage level is 110kV, supporting bidirectional power flow.

[0089] 2.1. The data acquisition module achieves real-time acquisition of multi-source data through "edge computing gateway + 5G communication". The acquired data includes electricity / heat / cooling load data for each user within the area, operating parameters of energy supply station equipment, renewable energy output data, power grid gateway power and price, natural gas supply pressure and price, and carbon emission factors (where the grid purchase factor is 0.55 kgCO2 / kWh and the natural gas factor is 2.05 kgCO2 / m³). 3 In the preprocessing stage, Kalman filtering is used to filter high-frequency fluctuations in wind power output, and the mean imputation method for adjacent time periods is used to process short-term missing data. Finally, a dataset with a 15-minute time scale (96 time segments per day) is constructed.

[0090] 2.2. Multi-objective energy-carbon synergistic optimization model and algorithm optimization: Based on Example 1, the multi-objective energy-carbon co-optimization model adds "power grid flow constraints" (requiring the regional 110kV threshold power to be ≤40MW) and "heat network / cold network hydraulic constraints" (stipulating that the pipeline pressure loss is ≤0.1MPa); the parameters of the information optimization Falcon algorithm are adjusted as follows: population size N=60, maximum number of iterations K=180, and inertia factor. The initial value is set to 0.95, which is the parameter for information interaction. During the algorithm iteration process, for the scenario of "coordination between regional power supply stations", the output of the two power supply stations is complemented by an information exchange network. For example, when the CCHP of power supply station No. 1 fails, power supply station No. 2 obtains the fault information through the network and quickly increases its output by 3MW to ensure load supply.

[0091] 2.3. Implementation Results: The implementation effects of the method of this invention in the regional energy internet are as follows: the typical daily total economic cost is 285,000 yuan (7.1% lower than PSO and 10.3% lower than GA), the total carbon emissions are 182.5 tCO2 (9.2% lower than PSO and 12.1% lower than GA), and the energy supply reliability is 99.78% (0.25 percentage points higher than PSO and 0.4 percentage points higher than GA), which fully verifies the applicability of this invention in large-scale integrated energy systems.

[0092] The two embodiments described above demonstrate that the integrated energy system energy-carbon synergistic optimization method of the present invention can achieve multi-objective synergistic optimization of "economy, low carbon emissions, and energy supply reliability" for integrated energy systems of different scales and scenarios. Compared with traditional intelligent optimization algorithms, it has significant advantages in solution accuracy, convergence speed, and robustness. The core innovation of the present invention lies in the introduction of the information optimization falcon algorithm. Through the deep integration of information interaction, dynamic feedback, and local development, it effectively breaks through the solution bottleneck of high-dimensional, multi-coupled, and multi-constrained optimization problems of integrated energy systems, providing a feasible technical solution for intelligent and low-carbon scheduling of energy systems in practical engineering.

[0093] In practical applications, the algorithm parameters (such as population size, number of iterations, and target weights) and model constraints (such as new equipment capacity constraints and hydrogen energy supply and demand balance constraints) can be flexibly adjusted according to the specific scale of the integrated energy system (such as park level or regional level), equipment configuration (such as whether it includes new equipment such as hydrogen energy storage and heat pumps), and target priorities (such as industrial parks focusing on economic efficiency and municipal areas focusing on low carbon emissions), so as to further expand the application scope of this invention.

[0094] The integrated energy system carbon-coordinated scheduling device provided by the present invention is described below. The integrated energy system carbon-coordinated scheduling device described below and the integrated energy system carbon-coordinated scheduling method described above can be referred to in correspondence.

[0095] The integrated energy system carbon coordination scheduling device provided by this invention refers to... Figure 6 As shown, it includes the following modules: Modeling module 210 is used to construct a multi-objective energy-carbon synergistic optimization model based on multi-source energy operation data and carbon emission factors of integrated energy systems. Optimization module 220 is used to solve the multi-objective energy-carbon co-optimization model using the information optimization falcon algorithm to obtain the Pareto optimal solution set; The scheduling module 230 is used to generate a scheduling scheme for the integrated energy system based on the Pareto optimal solution set.

[0096] The modeling module, based on the aforementioned multi-source energy operation data and carbon emission factors, constructs a multi-objective energy-carbon synergistic optimization model, comprehensively considering economic efficiency, low-carbon emissions, and energy supply reliability constraints. The optimization module employs the Information Optimization Falcon Algorithm, simulating the information exchange and rapid response behavior of a falcon during hunting to achieve global search and local development of the multi-objective scheduling problem. A dynamic information feedback mechanism improves convergence speed and solution accuracy. The scheduling module generates the optimal operating scheme for the multi-energy system based on the optimization results and issues it for execution. This invention utilizes the group collaboration and flexible adaptability of the Information Optimization Falcon Algorithm to effectively avoid the problem of traditional optimization methods easily getting trapped in local optima, achieving synergistic optimization of the integrated energy system between economic benefits and low carbon emissions. It possesses strong intelligent, adaptable, and engineering application value.

[0097] Figure 7 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 7 As shown, the electronic device may include: a processor 310, a communication interface 320, a memory 330, and a communication bus 340, wherein the processor 310, the communication interface 320, and the memory 330 communicate with each other via the communication bus 340. The processor 310 can call logical instructions in the memory 330 to execute a comprehensive energy system energy-carbon coordinated scheduling method, which includes: Based on the multi-source energy operation data and carbon emission factors of the integrated energy system, a multi-objective energy-carbon synergistic optimization model is constructed. The multi-objective energy-carbon co-optimization model was solved using the information-optimized Falcon algorithm to obtain the Pareto optimal solution set; Based on the Pareto optimal solution set, a scheduling scheme for the integrated energy system is generated.

[0098] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0099] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program that can be stored on a non-transitory computer-readable storage medium, wherein when the computer program is executed by a processor, the computer is able to execute the integrated energy system energy-carbon coordinated scheduling method provided by the above methods, the method comprising: Based on the multi-source energy operation data and carbon emission factors of the integrated energy system, a multi-objective energy-carbon synergistic optimization model is constructed. The multi-objective energy-carbon co-optimization model was solved using the information-optimized Falcon algorithm to obtain the Pareto optimal solution set; Based on the Pareto optimal solution set, a scheduling scheme for the integrated energy system is generated.

[0100] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the integrated energy system energy-carbon coordinated scheduling method provided by the above methods, the method comprising: Based on the multi-source energy operation data and carbon emission factors of the integrated energy system, a multi-objective energy-carbon synergistic optimization model is constructed. The multi-objective energy-carbon co-optimization model was solved using the information-optimized Falcon algorithm to obtain the Pareto optimal solution set; Based on the Pareto optimal solution set, a scheduling scheme for the integrated energy system is generated.

[0101] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0102] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0103] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for coordinated energy and carbon scheduling in an integrated energy system, characterized in that, include: Based on the multi-source energy operation data and carbon emission factors of the integrated energy system, a multi-objective energy-carbon synergistic optimization model is constructed. The multi-objective energy-carbon co-optimization model was solved using the information-optimized Falcon algorithm to obtain the Pareto optimal solution set; Based on the Pareto optimal solution set, a scheduling scheme for the integrated energy system is generated.

2. The integrated energy system carbon-coordinated scheduling method according to claim 1, characterized in that, The multi-objective energy-carbon synergistic optimization model is constructed based on the multi-source energy operation data and carbon emission factors of the integrated energy system, including: The multi-source energy operation data and carbon emission factors of the integrated energy system are used as inputs; Define an objective function, which includes at least the following objectives: minimizing economic costs, minimizing carbon emissions, and maximizing energy supply reliability; wherein, economic costs include electricity purchase costs, fuel costs, and equipment operation and maintenance costs; carbon emissions are calculated based on the carbon emission factors and consumption of various energy sources; and energy supply reliability is measured based on the ratio of the load deficit of electrical energy, heat energy, and cooling energy to the total load. The constraints are set, including at least energy balance constraints, coupling constraints, equipment capacity constraints, and ramp constraints. Energy balance constraints include the real-time supply and demand balance of electrical, thermal, and cooling energy. Coupling constraints include the ratio of electrical and thermal output of combined cooling, heating, and power (CCHP) units, and the ratio of thermal and cooling output of absorption chillers. Equipment capacity constraints include the upper and lower limits of output or energy storage for each energy device. Ramp constraints include restrictions on the rate of change of output of each energy device in adjacent time periods. By combining the objective function and the constraints, the multi-objective energy-carbon collaborative optimization model is obtained.

3. The integrated energy system carbon-coordinated scheduling method according to claim 1, characterized in that, Before solving the multi-objective energy-carbon co-optimization model using the Information Optimization Falcon Algorithm, the method further includes: The population size, maximum number of iterations, inertia factor, and information interaction parameters of the information optimization falcon algorithm are set, and an initial population is generated based on the decision variable range of the multi-objective energy-carbon collaborative optimization model. Calculate the information entropy of the initial population. If the information entropy is lower than a preset threshold, perform replacement or perturbation operations on some individuals to obtain an enhanced population. An information exchange network is established in the enhanced population. During the algorithm iteration process, the information exchange network is used to provide each individual with the optimal location information of its neighboring individuals. The neighboring individuals refer to other individuals that are directly connected to the individual through the information exchange network.

4. The integrated energy system carbon-coordinated scheduling method according to claim 3, characterized in that, Each individual in the information-optimized falcon algorithm corresponds to a set of candidate scheduling schemes in the multi-objective energy-carbon collaborative optimization model; The information-optimized Falcon algorithm is used to solve the multi-objective energy-carbon co-optimization model to obtain the Pareto optimal solution set, including: In the k-th iteration, the position of the individual after information exchange is determined based on the current position of the individual, the global optimal position of the individual in the current iteration, and the optimal position of the neighboring individuals obtained through the information exchange network. The position of the individual after the information exchange, the worst individual position in the current iteration, and a random number based on the Levy distribution are used to determine the position of the individual after the jump. Based on the individual position after the jump, the elite individual position in the current iteration, and a random number following a standard normal distribution, the individual position after local development is determined, where the elite individual position is the average position of the most fit individuals in the current population. Based on the convergence trend of the current iteration process, the inertia factor and information interaction parameters of the information-optimized falcon algorithm are adaptively adjusted to obtain the updated algorithm parameters. Using the individual position after local development as the initial position of the individual in the next iteration, and using the updated algorithm parameters as the initial parameters for the next iteration, perform the (k+1)th iteration until the iteration termination condition is met, and output the Pareto optimal solution set.

5. The integrated energy system carbon-coordinated scheduling method according to claim 2, characterized in that, The step of generating a scheduling scheme for the integrated energy system based on the Pareto optimal solution set includes: Using the analytic hierarchy process, we determined the corresponding weights for economic cost, carbon emission, and energy supply reliability for the three indicators: economic cost, carbon emission, and energy supply reliability. Based on the economic cost weight, carbon emission weight, and energy supply reliability weight, calculate the comprehensive evaluation value of each solution in the Pareto optimal solution set; The solution with the highest comprehensive evaluation value is selected from the Pareto optimal solution set as the final scheduling scheme.

6. The integrated energy system carbon coordination scheduling method according to any one of claims 1-5, characterized in that, The method further includes: The generated scheduling scheme is then distributed to the energy equipment of the integrated energy system for execution. Collect actual operating data after scheduling execution, and use the actual operating data as input data for constructing the multi-objective energy-carbon synergistic optimization model in the next round of optimization.

7. A comprehensive energy system carbon-coordinated dispatching device, characterized in that, include: The modeling module is used to construct a multi-objective energy-carbon synergistic optimization model based on multi-source energy operation data and carbon emission factors of integrated energy systems. The optimization module is used to solve the multi-objective energy-carbon co-optimization model using the information optimization falcon algorithm to obtain the Pareto optimal solution set; The scheduling module is used to generate a scheduling scheme for the integrated energy system based on the Pareto optimal solution set.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the integrated energy system energy and carbon coordinated scheduling method as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the integrated energy system energy-carbon coordinated scheduling method as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the integrated energy system energy-carbon coordinated scheduling method as described in any one of claims 1 to 6.

Images