Optimized scheduling method for park integrated energy system based on multi-objective white whale algorithm
By applying the non-dominant sorting beluga optimization algorithm in the integrated energy system of industrial parks, a multi-objective optimization model is built, and the problems of multi-energy coupling and multi-time scale scheduling are solved, and efficient management and scheduling of electricity, heat and cold energy are realized, reducing the efficiency of The operating costs and environmental impacts are applicable to industrial parks of different sizes.
Patent Information
- Application Number
- CN202411129403.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-16
- Publication Date
- 2024-11-19
AI Technical Summary
The integrated energy system in industrial parks has difficulties in multi-energy coupling, multi-time scale scheduling, multi-objective optimization and uncertainty processing. Traditional algorithms are inefficient in computing and are prone to local optimization, making it difficult to effectively manage and dispatch electricity, heat, cold, etc. Multiple energy sources.
Using the optimization scheduling method based on the non-dominant sorting beluga whale optimization algorithm (NSBWO), a comprehensive energy system model including electrical heating, energy storage, and micro-energy sources is constructed. Through the multi-objective optimization model and intelligent scheduling algorithm, combined with the price demand response mechanism, Optimize operational costs and environmental impacts to achieve multi-time scale scheduling.
It improves energy utilization efficiency, reduces operating costs and environmental impact, enhances the system's ability to adapt to uncertainty, and achieves dual optimization of economic and environmental friendliness, suitable for industrial parks of different sizes and types.
Smart Images

Figure CN118982204A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of optimization and scheduling of integrated energy systems, and in particular to an optimization and scheduling method applied to industrial park integrated energy systems (PIES). The method uses an advanced non-dominated sorting white whale optimization algorithm (NSBWO) to achieve effective management and scheduling of multiple energy forms in industrial parks, so as to achieve the dual optimization goals of operating costs and environmental impact. Background Art
[0002] With the rapid development of industrialization and urbanization, industrial parks, as important carriers of industrial production, have seen an increasing energy consumption. Traditional energy systems are often based on a single energy supply, which is difficult to meet the industrial park's demand for multiple energy sources such as electricity, heat, and cooling. In addition, single energy systems have problems such as low energy utilization efficiency, poor scheduling flexibility, and large environmental impact. In order to meet these challenges, the concept of Integrated Energy System (IES) came into being.
[0003] An integrated energy system is a system that integrates multiple energy forms such as electricity, heat, and cold energy. Through complementary use of energy and optimized scheduling, it improves energy efficiency, reduces operating costs, and reduces environmental pollution. However, the optimized scheduling of integrated energy systems faces many technical challenges:
[0004] Multi-energy coupling characteristics: There are complex coupling relationships between different energy forms in the integrated energy system, such as combined heat and power generation, combined heat and power generation, etc., which require precise modeling and analysis.
[0005] Multi-time scale scheduling: The energy demand of industrial parks has obvious time-varying characteristics, and it is necessary to consider scheduling strategies at different time scales such as day-ahead scheduling and intraday scheduling.
[0006] Multi-objective optimization problem: Optimal scheduling requires considering multiple objectives such as operating costs, environmental impact, and system stability. These objectives often conflict with each other and require a comprehensive trade-off.
[0007] Uncertainty handling: The volatility and uncertainty of renewable energy sources such as wind and solar energy pose additional challenges to the optimal dispatch of the system.
[0008] Intelligent scheduling algorithm: Traditional optimization algorithms have problems such as low computational efficiency and easy falling into local optimality when dealing with large-scale, multi-objective, and strongly constrained complex systems.
[0009] In response to the above problems, researchers have carried out a series of research work and proposed a variety of optimization scheduling methods and algorithms, such as genetic algorithm, particle swarm optimization algorithm, simulated annealing algorithm, etc. However, these methods still have certain limitations when dealing with the optimization scheduling problem of integrated energy systems in actual industrial parks.
[0010] Based on the above background technology, the present invention proposes an optimization scheduling method based on the non-dominated sorting Beluga optimization algorithm (NSBWO) for the specific needs of optimizing the scheduling of the integrated energy system in industrial parks. This method comprehensively considers the multi-energy coupling characteristics of the system, multi-time scale scheduling requirements, multi-objective optimization problems, uncertainty processing, and the design of intelligent scheduling algorithms, aiming to achieve efficient, economical, and environmentally friendly operation of the integrated energy system in industrial parks.
[0011] After searching, the application publication number CN116468168A, a distributed power multi-objective hierarchical planning method based on the improved Beluga optimization algorithm, includes constructing a distributed power multi-objective hierarchical planning model, and proposing a solution method based on the improved Beluga optimization algorithm. The output power of the distributed power source is uncertain, and the load-side demand response will also affect the planning results of the distributed power source. Therefore, the present invention aims to solve the optimal planning problem of the distributed power source access distribution network considering the load-side demand response and the uncertainty of the distributed power output power, so as to improve the economic benefits, power quality and energy utilization efficiency of the distribution network. First, for the multi-objective, multi-constrained, nonlinear distributed power planning problem, an improved Beluga optimization algorithm with excellent performance is proposed, and the effectiveness of the algorithm and the superiority of its performance are verified. Secondly, a distributed power multi-objective hierarchical planning model is established. Finally, the effectiveness of the method of the present invention is verified in a classic test system.
[0012] The patent mainly focuses on the planning of distributed power sources and does not cover the coordinated optimization of multiple energy forms in the integrated energy system. Although the White Whale optimization algorithm has been improved, there may be room for improvement in multi-objective optimization, especially the balance between operating costs and environmental impacts. The modeling of other energy forms such as heat and cold energy in the integrated energy system is not comprehensive enough.
[0013] This patent mainly proposes an optimization scheduling method for the integrated energy system of the park that comprehensively considers multiple energy forms such as electricity, heat, and cooling. Through the NSBWO algorithm, it optimizes operating costs and environmental impacts at the same time, providing a more comprehensive multi-objective optimization solution. A more comprehensive integrated energy system model is established, including electric heating, energy storage, micro energy sources, etc., which improves the practicality and accuracy of the model. The day-ahead scheduling and intraday scheduling strategies are clearly proposed to better adapt to the time-varying characteristics of energy demand. Summary of the invention
[0014] The present invention aims to solve the above problems of the prior art. A method for optimizing the scheduling of a park integrated energy system based on a multi-objective Beluga algorithm is proposed. The technical solution of the present invention is as follows:
[0015] An optimization scheduling method for a park integrated energy system based on a multi-objective Beluga algorithm comprises the following steps:
[0016] Establish a PIES model: Construct a PIES model of an industrial park integrated energy system that includes electric heating equipment, energy storage equipment and multiple micro-energy sources;
[0017] Construct a multi-objective optimization model: introduce a price-based demand response (PDR) mechanism and establish a multi-objective optimization model that considers a two-day time scale, with the goal of minimizing operating costs and environmental costs;
[0018] Initialize NSBWO algorithm parameters: Set the initial parameters of the non-dominated sorting White Whale optimization algorithm NSBWO, including population size, number of iterations, crossover rate, and mutation rate;
[0019] Population initialization: Use the good point set method GPS to initialize the initial population of the NSBWO algorithm with uniform distribution to improve the algorithm's search efficiency and global optimization capabilities;
[0020] Non-dominated sorting and fitness calculation: Perform non-dominated sorting on the individuals in the population and calculate the fitness value of each individual to ensure the diversity of the population and the quality of the optimization solution;
[0021] Generate new individuals: Apply crossover and mutation operations to generate new individuals, enhance the diversity of the population, improve the algorithm's search ability and avoid falling into local optimality;
[0022] Update the population and optimize the solution set: Update the population and optimize the solution set according to the fitness and non-dominated sorting results, and maintain the superiority of the population through the elite selection strategy;
[0023] Termination condition judgment: Determine whether the algorithm meets the preset termination conditions, such as reaching the maximum number of iterations or the quality of the solution reaching the expected goal; if satisfied, output the optimal scheduling strategy; if not satisfied, return to continue iteration.
[0024] Furthermore, the detailed model of the PIES is established, including an electric heater (EB), a battery storage system (BESS), a wind turbine (WT), a photovoltaic system (PV), a gas turbine (GT) and a waste heat boiler (WHB). The GT and WHB together form a CHP. The mathematical model of the combined heat and power unit CHP is described as follows:
[0025]
[0026] Where ηMT (t), P MT (t) is the gas turbine GT operating efficiency and GT output power; C MT,OM (t),C MT,FU (t),C MT,EN (t) are the operation and maintenance cost, fuel cost and emission cost of GT, respectively. is the natural gas fuel cost coefficient, s,c PO ,γ MT,s are respectively the type of exhaust gas, gas treatment cost coefficient, and gas exhaust volume, K OM is the coefficient of GT operation and maintenance cost; s and n represent the emission gas types respectively;
[0027] P WHB (t) = ρ MT η WHB P MT (t)
[0028] Where P WHB is the heat generation power of WHB. MT is the GT thermal efficiency, η WHB is the conversion efficiency of WHB to high temperature flue gas.
[0029] Furthermore, the multi-objective optimization model is constructed: a price-based demand response (PDR) mechanism is introduced to establish a multi-objective optimization model that considers a two-day time scale, with the goal of minimizing operating costs and environmental costs, specifically including:
[0030] Including operating costs F1 and environmental costs F2, use the following formula:
[0031]
[0032]
[0033] In the formula, C MT-WHB (t)C bess (t)C hst (t)C W (t)C PV (t)C g (t) is the total operation and maintenance cost of GT, WHB, BESS, TST, WT and PV; K bess , K hst , K W , K PV are the maintenance cost coefficients of BESS, TST, WT and PV respectively; C g (t) is the cost of purchasing electricity from the power grid; P g (t) is the power of the main power grid; S buy (t) is the electricity purchase price of the power grid at time t;
[0034] Minimize environmental costs:
[0035] F2=C MT,EN +C g,EN
[0036]
[0037] C g,EN is the environmental cost of the main power grid, and γg,s is the gas emission per unit of power generation in the power grid.
[0038] Furthermore, the optimization model includes the following constraints:
[0039] Power balance constraints:
[0040] P W (t)+P PV (t)+P MT (t)+P bess (t)+P g (t)+P PEH (t) = P load (t)
[0041] Where P load (t) is the power load at time t;
[0042] Thermal balance constraints:
[0043] P WHB (t)+P hst (t)+H PEH (t) = H load (t)
[0044] In the formula, H load (t) is the heat load at time t; P WHB (t), P hst (t), H PEH (t) represents the heat generation power of WHB, the charge and discharge power of TST, and the heat generation of EB respectively.
[0045] Gas turbine output power and slope power constraints:
[0046]
[0047] In the formula, is the upper and lower limits of GT output power; P step is the maximum ramp power of the GT in a given period of time;
[0048] Grid output power constraints:
[0049] 0≤P g (t)≤Pmax
[0050] Where P max It is the upper limit of the amount of electricity that the park can purchase from the main power grid to prevent excessive fluctuations in the power grid from bringing new safety risks to the operation of the power grid;
[0051] Battery Energy Storage System (BESS) capacity and charging and discharging power constraints:
[0052]
[0053] In the formula, is the upper limit of the charging and discharging power of BESS; It is the upper and lower limits of battery capacity to prevent excessive discharge depth from affecting battery life;
[0054] Thermal storage system (TST) capacity and heat exchange power constraints:
[0055]
[0056] In the formula, is the upper limit of TST heat storage and release power, To ensure the upper and lower limits of the tank capacity to ensure safe operation of TST.
[0057] Furthermore, the electric boiler (EB) generates heat by using electricity and participates in the heating in the industrial park. The mathematical model of EB can be described as:
[0058] H PEH (t) = λ con λ los P PEH
[0059] Where P PEH , H PEH are the power consumption and heat generation of EB, respectively, con λ los are the energy conversion coefficient and heat transfer loss coefficient of the EB plate respectively.
[0060] Furthermore, the mathematical model of the battery energy storage system (BESS) can be described as:
[0061]
[0062] In the formula, SOC E (t), SOC E (t-1) are the remaining capacities of BESS at time t and t-1 respectively; P bess (t) is the charging and discharging power of BESS, where P bess (t)>0 indicates discharge, P bess (t)≤0 means charging is the energy conversion efficiency of BESS during the charging and discharging process.
[0063] Furthermore, the mathematical model of the thermal storage tank (TST) can be described as follows:
[0064]
[0065] In the formula, SOC H (t), SOC H (t-1) is the residual heat of TST at time t and t-1; P hst (t) is the charge and discharge power of TST, where P hst (t)≤0 means heat storage, P hst (t)>0 indicates exothermicity; Indicates the heat release and heat storage conversion efficiency of TST.
[0066] Furthermore, the NSBWO algorithm specifically includes the following steps:
[0067] Initialize the population and set algorithm parameters, including population size and number of iterations; use the good point set method (GPS) to initialize the population to ensure the diversity of the initial solution;
[0068] The BWO algorithm consists of three phases: exploration, exploitation, and whale fall. The BWO algorithm uses the balance factor Bf to promote the transition of the population from the exploration phase to the exploitation phase, which is expressed as:
[0069]
[0070] Among them, t is the current iteration count, T is the maximum number of iterations, and B0 changes randomly in the range of (0,1) at each iteration. f >0.5, the population is in the exploration stage. f When ≤0.5, the population is in the development stage. As the number of iterations t increases, B f The fluctuation range of gradually decreases from (0,1) to (0,0.5); the probability of the development stage increases with the number of iterations;
[0071] During the exploration phase of BWO, the position of the search agent is determined by the swimming of a pair of beluga whales, and the position of the beluga whales is updated as follows:
[0072]
[0073] in, is the new position of the ith beluga whale in the jth dimension, Pj is randomly selected from the dth dimension, is the position of the i-th beluga whale in the j-th dimension, and is the current 1 position of the ith and rth white whales, where r represents a randomly selected whale, and r1 and r2 are random numbers from 0 to 1 used to enhance the random operator in the exploration phase;
[0074] The BWO algorithm introduces the Levy flight strategy during the development phase, and its mathematical model is as follows:
[0075]
[0076] in, and are the current positions of the i-th beluga whale and a random beluga whale, is the new position of the i-th beluga whale, is the optimal position between white whales, r3 and r4 are random numbers ranging from 0 to 1, and c1 is the random jump intensity used to measure the intensity of Levi's flight;
[0077] Using the current position of the white whale and the step length of the white whale's descent to establish the updated position, the mathematical model is expressed as follows:
[0078]
[0079] Among them, r5, r6, r7 are random numbers between 0 and 1, x step represents the step size of whale falls, c2 is a step size factor, which is related to the probability of whale falls and the population size;
[0080] The probability of a whale falling Wf can be expressed as:
[0081]
[0082] Perform non-dominated sorting and fitness calculation to calculate the fitness of each individual:
[0083] φ(x)=f′(x)+r·M′(x)
[0084] Where f′(x) is the normalized objective function value of individual x, M′(x) is the normalized value of the constraint violation degree of individual x, and r is the penalty coefficient set according to the specific problem;
[0085] Apply crossover and mutation operations to generate new individuals to enhance the diversity of the population; update the population and select individuals with higher non-dominated sorting and higher fitness. Repeat the steps of non-dominated sorting, fitness calculation, new individual generation and population update until the termination condition is met.
[0086] Furthermore, the normalization process can be expressed as:
[0087]
[0088] Among them, Gi (x), m are inequality constraints and their number, h i (x), p are equality constraints and their numbers, ε is a positive number approaching zero, and ri is the adaptive penalty coefficient;
[0089]
[0090] Among them, fit-mark is the feasibility index, where 1 represents feasibility and 0 represents obsolescence;
[0091]
[0092] The advantages and beneficial effects of the present invention are as follows:
[0093] The present invention achieves efficient use of energy in industrial parks by establishing a comprehensive energy system model to accurately simulate the conversion and coupling relationship of multiple energy forms such as electricity, heat, and cold. Through a multi-objective optimization model, the present invention minimizes operating costs, including costs for energy purchase, equipment operation and maintenance, thereby saving expenses for industrial parks. The present invention takes environmental costs into account in the optimization scheduling, and reduces greenhouse gas emissions and other environmental impacts by reducing the use of fossil energy and optimizing the scheduling of renewable energy.
[0094] By adopting a multi-time scale scheduling strategy, the present invention can flexibly respond to fluctuations in energy demand and supply, and improve the system's adaptability to uncertainty. The present invention balances operating costs and environmental impacts through the NSBWO algorithm, achieving dual optimization of economy and environmental friendliness.
[0095] The optimization scheduling method of the present invention is applicable to industrial parks of different sizes and types, and has good versatility and adaptability. The present invention can formulate the optimal equipment operation strategy based on real-time data and prediction models, extend the service life of equipment, and reduce the failure rate.
[0096] The present invention complies with the national policy orientation of energy conservation, emission reduction and sustainable development, and helps industrial parks to meet the requirements of environmental protection laws and regulations. By implementing the above technical solution, the present invention not only improves the dispatching efficiency and economy of the comprehensive energy system of the industrial park, but also makes positive contributions to environmental protection and sustainable development.
[0097] The innovation of the present invention is that a method for optimizing the scheduling of an industrial park integrated energy system (PIES) based on the non-dominated sorting Beluga optimization algorithm (NSBWO) is proposed. This method constructs a multi-objective optimization model that simultaneously considers operating costs and environmental impacts by accurately modeling multiple energy forms and their conversion and coupling relationships. The NSBWO algorithm combines the advantages of NSGA-II and BWO, showing stronger search capabilities and faster convergence speed, and effectively solves the optimization problem of the industrial park energy system. In addition, the design of multi-time scale scheduling strategies improves the system's adaptability to the time-varying characteristics of energy demand, while the development of intelligent control and user interaction systems realizes real-time monitoring and adjustment of the optimization scheduling strategy. These innovations not only improve the scheduling efficiency and economy of the system, but also reduce environmental impact, which is in line with the policy orientation of sustainable development. The cleverness lies in the flexibility of algorithm fusion and multi-time scale scheduling, as well as the intelligent processing of uncertainty, which enhances the reliability of the system and the participation of users. BRIEF DESCRIPTION OF THE DRAWINGS
[0098] Figure 1 The present invention provides the main components and energy flow paths of the industrial park integrated energy system in the preferred embodiment.
[0099] Figure 2 The flowchart of the NSBWO algorithm is described, including the steps of initialization, non-dominated sorting, fitness calculation, new individual generation and population update. DETAILED DESCRIPTION
[0100] The following will describe the technical solutions in the embodiments of the present invention in detail in conjunction with the accompanying drawings in the embodiments of the present invention. The described embodiments are only part of the embodiments of the present invention.
[0101] The technical solution of the present invention to solve the above technical problems is:
[0102] First, the system is modeled: a detailed model of PIES is established, including electric heaters (EB), energy storage systems (BESS), wind turbines (WT), photovoltaic systems (PV), gas turbines (GT), and waste heat boilers (WHB). GT and WHB together form CHP. During the GT power generation operation, high-temperature flue gas is discharged and recovered by WHB, and then converted into steam for heating. The mathematical model of this CHP is described as follows:
[0103]
[0104] Where η MT (t), P MT (t) is the GT operating efficiency and GT output power. MT,OM (t), C MT,FU (t), C MT,EN(t) are the operation and maintenance cost, fuel cost and emission cost of GT respectively is the natural gas fuel cost coefficient. POγMT,s K is the type of exhaust gas, gas treatment cost coefficient, and gas exhaust volume. OM is the coefficient of GT operation and maintenance cost.
[0105] P WHB (t) = ρ MT η WHB P MT (t)
[0106] Where P WHB is the heat generation power of WHB. MT is the GT thermal efficiency, η WHB is the conversion efficiency of WHB to high temperature flue gas.
[0107] Furthermore, electric boilers (EB) participate in the heating supply in industrial parks by generating heat through electricity. Compared with WHB, EB has a faster response speed and lower environmental cost. The mathematical model of EB can be described as:
[0108] H PEH (t) = λ con λ los P PEH
[0109] Where P PEH , H PEH is the power consumption and heat generation of EB. con λ los are the energy conversion coefficient and heat transfer loss coefficient of the EB plate respectively.
[0110] Furthermore, the battery energy storage system (BESS) assists in peak load regulation and valley compensation through charging and discharging, reduces the impact of renewable energy generation fluctuations on the power grid, and ensures the stable and safe operation of the microgrid system. The mathematical model of BESS can be described as:
[0111]
[0112] In the formula, SOC E (t), SOC E (t-1) are the remaining capacities of BESS at time t and t-1 respectively; P bess (t) is the charging and discharging power of BESS, where P bess (t)>0 indicates discharge, P bess (t)≤0 means charging; is the energy conversion efficiency of BESS during the charging and discharging process.
[0113] Further, thermal storage tank (TST): When there is excess heat generated in the industrial park, TST starts working by storing heat to avoid energy waste. TST releases heat during the peak period of heating demand to alleviate the heating pressure in the industrial park. The mathematical model of the thermal storage tank can be described as follows:
[0114]
[0115] In the formula, SOC H (t), SOC H (t-1) is the residual heat of TST at time t and t-1; P hst (t) is the charge and discharge power of TST, where P hst (t)≤0 means heat storage, P hst (t)>0 indicates exothermicity; Indicates the heat release and heat storage conversion efficiency of TST.
[0116] Furthermore, the multi-objective optimization model is constructed: the optimization objective function is defined, including the operating cost F1 and the environmental cost F2, using the following formula:
[0117]
[0118]
[0119] In the formula, C MT-WHB (t)C bess (t)C hst (t)C W (t)C PV (t)C g (t) is the total operation and maintenance cost of GT, WHB, BESS, TST, WT and PV. bess , K hst , K W , K PV are the maintenance cost coefficients of BESS, TST, WT and PV respectively. g (t) is the cost of purchasing electricity from the power grid. g (t) is the power of the main power grid. S buy (t) is the power purchase price of the power grid at time t.
[0120] Furthermore, environmental costs are minimized:
[0121] F2=C MT,EN +C g,EN
[0122]
[0123] C g,ENis the environmental cost of the main power grid. γg,s is the gas emission per unit of power generation in the power grid.
[0124] Furthermore, the NSBWO algorithm is implemented by: initializing the population, setting algorithm parameters such as population size and number of iterations, and using the good point set method (GPS) to initialize the population to ensure the diversity of the initial solution.
[0125] The BWO algorithm is inspired by the life behavior of white whales and contains three stages: exploration, exploitation, and whale fall. The BWO algorithm uses the balance factor Bf to promote the transition of the population from the exploration stage to the exploitation stage, which can be expressed as:
[0126]
[0127] Where t is the current iteration count, T is the maximum number of iterations, and B0 changes randomly in the range of (0,1) at each iteration. f >0.5, the population is in the exploration stage. f When ≤0.5, the population is in the development stage. As the number of iterations t increases, B f The fluctuation range of decreases from (0,1) to (0,0.5). The probability of the development stage increases with the number of iterations.
[0128] Furthermore, during the exploration phase of BWO, the position of the search agent is determined by the movement of a pair of beluga whales. The position of the beluga whales is updated as follows:
[0129]
[0130] in, is the new position of the ith beluga whale in the jth dimension, Pj is randomly selected from the dth dimension, is the position of the i-th beluga whale in the j-th dimension, and are the current 1 positions of the ith and rth white whales (where r represents a randomly selected whale), and r1 and r2 are random numbers from 0 to 1 used to enhance the random operator in the exploration phase.
[0131] Furthermore, the BWO algorithm introduces the Levy flight strategy in the development stage to improve convergence, and its mathematical model is as follows:
[0132]
[0133] in, and are the current positions of the i-th beluga whale and a random beluga whale, is the new position of the i-th beluga whale, is the optimal position between the white whales, r3 and r4 are random numbers in the range of 0 to 1, and c1 is the random jump strength used to measure the strength of Levi's flight.
[0134] Furthermore, the updated position is established using the current position of the white whale and the step length of the white whale's descent. This mathematical model is expressed as follows:
[0135]
[0136] Among them, r5, r6, r7 are random numbers between 0 and 1. step represents the step size of whale falls. c2 is a step size factor, which is related to the probability of whale falls and the population size.
[0137] Furthermore, the probability of a whale falling Wf can be expressed as:
[0138]
[0139] Furthermore, non-dominated sorting and fitness calculation are performed to calculate the fitness of each individual:
[0140] φ(x)=f′(x)+r·M′(x)
[0141] Among them, f′(x) is the normalized objective function value of individual x, which reflects the performance of the individual in multiple objectives. M′(x) is the normalized value of the constraint violation degree of individual x, which is used to deal with the constraints in the optimization problem. r is a penalty coefficient set according to the specific problem, which is used to adjust the importance of the objective function value and the constraint violation value. The normalization process can be expressed as:
[0142]
[0143] Among them, G i (x), m are inequality constraints and their number, h i (x), p are equality constraints and their quantities, ε is a positive number approaching zero, and ri is the adaptive penalty coefficient.
[0144]
[0145] Among them, fit-mark is the feasibility index, where 1 represents feasibility and 0 represents obsolescence.
[0146]
[0147] Furthermore, crossover and mutation operations are applied to generate new individuals to enhance the diversity of the population. The population is updated and individuals with higher non-dominated sorting and higher fitness are selected. The steps of non-dominated sorting, fitness calculation, new individual generation and population update are repeated until the termination condition is met.
[0148] Furthermore, multi-time scale scheduling strategy: design day-ahead scheduling and intraday scheduling strategies, taking into account the changes in energy demand and supply at different time scales. Use forecast data and real-time data to dynamically adjust the balance of energy supply and demand to ensure the stability and economy of the system in different time periods.
[0149] Further, constraint processing: clarify power and thermal balance constraints, such as power balance:
[0150] P W (t)+P PV (t)+P MT (t)+P bess (t)+P g (t)+P PEH (t) = P load (t)
[0151] P load (t) is the power load at time t
[0152] Furthermore, the thermal balance constraint is:
[0153] P WHB (t)+P hst (t)+H PEH (t) = H load (t)
[0154] H load (t) is the heat load at time t
[0155] Furthermore, the gas turbine output power and slope power constraints are:
[0156]
[0157] in, P is the upper and lower limits of GT output power. step is the maximum ramp power of GT in a given time period. step The unit is kW / min and varies with the time interval on two different time scales.
[0158] Furthermore, the grid output power constraint is:
[0159] 0≤P g (t)≤P max
[0160] Among them, P max It is the upper limit of the amount of electricity that the park can purchase from the main power grid in order to prevent excessive fluctuations in the power grid from bringing new safety risks to the operation of the power grid.
[0161] Furthermore, the battery energy storage system (BESS) capacity and charging and discharging power constraints are:
[0162]
[0163] in, is the upper limit of the charging and discharging power of BESS. It is the upper and lower limits of battery capacity to prevent excessive discharge depth from affecting battery life.
[0164] Furthermore, the thermal storage system (TST) capacity and heat exchange power constraints are:
[0165]
[0166] in, It is the upper limit of TST heat storage and release power. To ensure the upper and lower limits of the tank capacity to ensure safe operation of TST.
[0167] Set the operating constraints of GT, BESS, WT, and PV, including minimum and maximum output limits, as well as charge and discharge power constraints.
[0168] Furthermore, an intelligent control unit is developed to achieve real-time monitoring and automatic adjustment of energy equipment to ensure that the system operates according to the optimized scheduling results. A user interaction interface is designed to enable users to input parameters, view scheduling results, and manage the system. The effectiveness of the NSBWO algorithm is verified through energy system data from an actual industrial park, including the convergence of the algorithm and the quality of the solution.
[0169] Compare the algorithm results with existing scheduling strategies to demonstrate optimization effects, such as cost savings and reduced environmental impact. Apply the optimization results to the scheduling of integrated energy systems in actual industrial parks to guide actual operations. Adjust energy equipment operations based on the optimization results to achieve cost savings and environmental benefits.
[0170] Implement a real-time monitoring system to collect operating data and environmental parameters of each energy device. According to the monitoring data and feedback, adjust the optimization model and scheduling strategy to adapt to the uncertainty and changes in system operation.
[0171] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions.
[0172] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0173] The above embodiments should be understood to be only used to illustrate the present invention and not to limit the protection scope of the present invention. After reading the contents of the present invention, technicians can make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.
Claims
1. An optimization scheduling method for the integrated energy system of a park based on the multi-objective Beluga algorithm. It is characterized in that The following steps are involved: Establish a PIES model: Construct a PIES model of an industrial park integrated energy system that includes electric heating equipment, energy storage equipment and multiple micro-energy sources; Construct a multi-objective optimization model: introduce a price-based demand response (PDR) mechanism and establish a multi-objective optimization model that considers a two-day time scale, with the goal of minimizing operating costs and environmental costs; Initialize NSBWO algorithm parameters: Set the initial parameters of the non-dominated sorting White Whale optimization algorithm NSBWO, including population size, number of iterations, crossover rate, and mutation rate; Population initialization: Use the good point set method GPS to initialize the initial population of the NSBWO algorithm with uniform distribution to improve the algorithm's search efficiency and global optimization capabilities; Non-dominated sorting and fitness calculation: Perform non-dominated sorting on the individuals in the population and calculate the fitness value of each individual to ensure the diversity of the population and the quality of the optimization solution; Generate new individuals: Apply crossover and mutation operations to generate new individuals, enhance the diversity of the population, improve the algorithm's search ability and avoid falling into local optimality; Update the population and optimize the solution set: Update the population and optimize the solution set according to the fitness and non-dominated sorting results, and maintain the superiority of the population through the elite selection strategy; Termination condition judgment: Determine whether the algorithm meets the preset termination conditions, such as reaching the maximum number of iterations or the quality of the solution reaching the expected goal; if satisfied, output the optimal scheduling strategy; if not satisfied, return to continue iteration.
2. According to claim 1, the optimization scheduling method of the park integrated energy system based on the multi-objective Beluga algorithm, It is characterized in that The detailed model of PIES is established, including electric heater (EB), energy storage system (BESS), wind turbine (WT), photovoltaic system (PV), gas turbine (GT) and waste heat boiler (WHB). GT and WHB together form CHP. The mathematical model of CHP of combined heat and power unit is described as follows: Where η MT (t), P MT (t) is the gas turbine GT operating efficiency and GT output power; C MT,OM (t),C MT,FU (t),C MT,EN (t) are the operation and maintenance cost, fuel cost and emission cost of GT, respectively. is the natural gas fuel cost coefficient, s,c PO ,γ MT,s are respectively the type of exhaust gas, gas treatment cost coefficient, and gas exhaust volume, K OM is the coefficient of GT operation and maintenance cost; s and n represent the type of released gas respectively; P WHB (t)=ρ MT or WHB P MT (t) Where P WHB is the heat generation power of WHB. MT is the GT thermal efficiency, η WHB is the conversion efficiency of WHB to high temperature flue gas.
3. The optimization scheduling method of the park integrated energy system based on the multi-objective Beluga algorithm according to claim 2, It is characterized in that The multi-objective optimization model is constructed by introducing a price-based demand response (PDR) mechanism and establishing a multi-objective optimization model that considers a two-day time scale, with the goal of minimizing operating costs and environmental costs, including: Including operating costs F1 and environmental costs F2, use the following formula: In the formula, C MT-WHB (t)C bess (t)C hst (t)C W (t)C PV (t)C g (t) is the total operation and maintenance cost of GT, WHB, BESS, TST, WT and PV; K bess , K hst , K W , K PV are the maintenance cost coefficients of BESS, TST, WT and PV respectively; C g (t) is the cost of purchasing electricity from the power grid; P g (t) is the power of the main power grid; S buy (t) is the electricity purchase price of the power grid at time t; Minimize environmental costs: F2=C MT,EN +C g,EN C g,EN is the environmental cost of the main power grid, and γg,s is the gas emission per unit of power generation in the power grid.
4. The optimization scheduling method for the integrated energy system of a park based on the multi-objective Beluga algorithm according to claim 2, It is characterized in that The following constraints are included in the optimization model: Power balance constraints: P W (t)+P PV (t)+P MT (t)+P bess (t)+P g (t)+P PEH (t)=P load (t) Where P load (t) is the power load at time t; Thermal balance constraints: P WHB (t)+P hst (t)+H PEH (t)=H load (t) In the formula, H load (t) is the heat load at time t; P WHB (t), P hst (t), H PEH (t) represents the heat generation power of WHB, the charge and discharge power of TST, and the heat generation of EB; Gas turbine output power and slope power constraints: In the formula, is the upper and lower limits of GT output power; P step is the maximum ramp power of the GT in a given period of time; Grid output power constraints: 0≤P g (t)≤P max Where P max It is the upper limit of the amount of electricity that the park can purchase from the main power grid to prevent excessive fluctuations in the power grid from bringing new safety risks to the operation of the power grid; Battery Energy Storage System (BESS) capacity and charging and discharging power constraints: In the formula, is the upper limit of the charging and discharging power of BESS; It is the upper and lower limits of battery capacity to prevent excessive discharge depth from affecting battery life; Thermal storage system (TST) capacity and heat exchange power constraints: In the formula, is the upper limit of TST heat storage and release power, To ensure the upper and lower limits of the tank capacity to ensure safe operation of TST.
5. The optimization scheduling method for the integrated energy system of a park based on the multi-objective Beluga algorithm according to claim 1, It is characterized in that Electric boilers (EB) generate heat by using electricity and participate in the heating in the industrial park. The mathematical model of EB can be described as: H PEH (t)=λ con l los P PEH Where P PEH , H PEH are the power consumption and heat generation of EB, respectively, con λ los are the energy conversion coefficient and heat transfer loss coefficient of the EB plate respectively.
6. The optimization scheduling method for the integrated energy system of a park based on the multi-objective Beluga algorithm according to claim 1, It is characterized in that The mathematical model of the battery energy storage system (BESS) can be described as: In the formula, SOC E (t), SOC E (t-1) are the remaining capacities of BESS at time t and t-1 respectively; P bess (t) is the charging and discharging power of BESS, where P bess (t)>0 indicates discharge, P bess (t)≤0 means charging; is the energy conversion efficiency of BESS during the charging and discharging process.
7. The optimization scheduling method for the integrated energy system of a park based on the multi-objective Beluga algorithm according to claim 1, It is characterized in that The mathematical model of the thermal storage tank (TST) can be described as follows: In the formula, SOC H (t), SOC H (t-1) is the residual heat of TST at time t and t-1; P hst (t) is the charge and discharge power of TST, where P hst (t)≤0 means heat storage, P hst (t)>0 indicates exothermicity; Indicates the heat release and heat storage conversion efficiency of TST.
8. The optimization scheduling method for the integrated energy system of a park based on the multi-objective Beluga algorithm according to claim 1, It is characterized in that The NSBWO algorithm specifically includes the following steps: Initialize the population and set algorithm parameters, including population size and number of iterations; use the good point set method (GPS) to initialize the population to ensure the diversity of the initial solution; The BWO algorithm consists of three phases: exploration, exploitation, and whale fall. The BWO algorithm uses the balance factor Bf to promote the transition of the population from the exploration phase to the exploitation phase, which is expressed as: Among them, t is the current iteration count, T is the maximum number of iterations, B 0 In each iteration, it changes randomly in the range of (0,1). f >0.5, the population is in the exploration stage. f When ≤0.5, the population is in the development stage. As the number of iterations t increases, B f The fluctuation range of gradually decreases from (0,1) to (0,0.5); the probability of the development stage increases with the number of iterations; During the exploration phase of BWO, the position of the search agent is determined by the swimming of a pair of beluga whales, and the position of the beluga whales is updated as follows: in, is the new position of the ith beluga whale in the jth dimension, Pj is randomly selected from the dth dimension, is the position of the i-th beluga whale in the j-th dimension, and is the current 1 position of the i-th and r-th beluga whales, where r represents a randomly selected whale and r 1 and r 2 is a random number between 0 and 1, used to enhance the random operator in the exploration phase; The BWO algorithm introduces the Levy flight strategy during the development phase, and its mathematical model is as follows: in, and are the current positions of the i-th beluga whale and a random beluga whale, is the new position of the i-th beluga whale, It is the best position among white whales. 3 and r 4 is a random number between 0 and 1, c 1 is the random jump intensity used to measure the intensity of Levi's flight; Using the current position of the white whale and the step length of the white whale's descent to establish the updated position, the mathematical model is expressed as follows: Among them, r 5 ,r 6 ,r 7 is a random number between 0 and 1, x step represents the step length of the whale fall, c 2 is a step size factor that is related to the probability of whale falls and the population size; The probability of a whale falling Wf can be expressed as: Perform non-dominated sorting and fitness calculation to calculate the fitness of each individual: φ(x)=f′(x)+r·M′(x) Where f′(x) is the normalized objective function value of individual x, M′(x) is the normalized value of the constraint violation degree of individual x, and r is the penalty coefficient set according to the specific problem; Apply crossover and mutation operations to generate new individuals to enhance the diversity of the population; update the population and select individuals with higher non-dominated sorting and higher fitness. Repeat the steps of non-dominated sorting, fitness calculation, new individual generation and population update until the termination condition is met.
9. The optimization scheduling method for the integrated energy system of a park based on the multi-objective Beluga algorithm according to claim 8, It is characterized in that The normalization process can be expressed as: Among them, G i (x), m are inequality constraints and their number, h i (x), p are equality constraints and their numbers, ε is a positive number approaching zero, and ri is the adaptive penalty coefficient; Among them, fit-mark is the feasibility index, where 1 represents feasibility and 0 represents obsolescence;
Citation Information
Cited By
Integrated energy system optimization method and device based on improved whale algorithm, and medium
CN119918844A
Multi-channel optical switch test method and system, and storage medium
CN122545071A
