Comprehensive energy optimization control method and system based on multi-objective optimization

Abstract

The invention discloses a comprehensive energy optimization control method and system based on multi-objective optimization, and the method comprises the steps: firstly constructing a comprehensive energy system structure model which comprises a power system model, a natural gas system model and a thermodynamic system model, and carrying out the mathematical modeling of typical equipment of each part of an energy system; constructing a population by taking parameters of the integrated energy system model as original positions of individuals, and optimizing the parameters in the integrated energy system model by taking a target function of the integrated energy system model as a fitness function and performing repeated iterative updating on the original positions of the individuals; an innovative solution is provided for multi-objective optimization of the comprehensive energy system, a new technical path is provided for the optimization problem of the comprehensive energy system, and theoretical support and practical reference are provided for energy conservation, emission reduction and green development in the energy field.

Description

Technical Field

[0001] The present invention belongs to the technical field of integrated energy systems, and in particular relates to an integrated energy optimization control method and system based on multi-objective optimization. Background Art

[0002] With the rapid development of human society in today's world, the demand for energy is increasing, which has led to a contradiction between supply and demand of resource utilization and the shortage of fuel resources. Through gradual exploration, people have realized that making full use of renewable resources can not only alleviate the contradiction between supply and demand, but also ease the contradiction between industrial development and environmental protection.

[0003] In recent years, the proposal of the Integrated Energy System (IES) has effectively improved energy utilization. IES includes multiple energy sources such as wind, heat, electricity, and gas. Energy diversification means that there will be scenarios where multiple energy sources will be developed and utilized in a coordinated manner. At the same time, it will also lead to the problem of coupling between multiple energy sources. For example, should electricity in a certain period of time be generated by wind energy or heat energy? How to couple them to maximize the utilization rate of the integrated energy system, minimize costs, and maximize benefits?

[0004] Wind energy is the main part of the current development of new energy; however, although wind power generation technology has become mature and has been invested in the power system in large quantities, there are still some problems in the connection and development of wind power, such as: there are obvious differences in the degree of regional wind power generation, and the problem of wind power abandonment is prominent (China Industry Information. Analysis of the Development Status and Development Prospects of China's Wind Power Industry in 2018 [EB / OL]. 2018-05-31 / 2019-1-6.). The literature (Salimi M, Ghasemi H, Adelpour M, Vaez-ZAdeh S. Optimal planning of energy hubs in interconnected energy systems: a case study for natural gas and electricity [J]. IET Generation Transmission & Distribution, 2015, 9 (8): 695-707.) proposed an energy center framework model for optimal design and large-scale interconnection, which is conducive to solving the problems of interconnected energy center planning and future energy system in integrated energy systems. Starting from the concept of energy hub, Wang Yi comprehensively considered the coordination of different forms of energy such as electricity, heat, and natural gas in multiple links such as production, transportation, and consumption (Wang Yi, Zhang Ning, Kang Chongqing. A review and prospect of research on optimization planning and operation of energy hubs in energy internet [J]. Proceedings of the Chinese Society of Electrical Engineering, 2015, 35(22): 5669-5681). For the above situation, the distribution system in the integrated energy system must be restructured to establish an energy coordination optimization model to make the urban energy system low-cost and efficient. Considering the impact of the natural gas system, before studying the electric-gas integrated energy system, a comprehensive system model must be established. In the existing literature, Song Nan considered the impact of the connection of the two energy networks on the mixed flow (Song Nan. Planning of the electric-gas integrated energy system considering P2G plants [D]. Shanghai Institute of Electric Power, 2020.), but did not fully consider the impact of the output fluctuation of the coupling equipment on the entire integrated energy system. At the same time, the existing technology is prone to fall into local optimality when optimizing multiple parameters of the integrated energy system, which makes the control scheme of the integrated energy system unable to achieve the optimality. Summary of the Invention

[0005] The purpose of the present invention is to provide a comprehensive energy optimization control method and system based on multi-objective optimization.

[0006] The present invention provides a comprehensive energy optimization control method based on multi-objective optimization, which includes:

[0007] Construct an integrated energy system model and set the objective function and constraints of the integrated energy system; construct a population using the integrated energy system model parameters as the original positions of individuals, and use the objective function of the integrated energy system model as the fitness function. Optimize the parameters in the integrated energy system model by iteratively updating the original positions of individuals multiple times; in one iteration, update the individual positions sequentially through the first, second, and third stages;

[0008] The first stage updates the position of the individual based on its original position; the second stage constructs the position of the threat and updates the position based on the threat; the third stage generates a random position by introducing the best individual position and the worst individual position in the current population and updates the position based on the random position; after the iteration termination condition is reached, the individual position with the smallest fitness is used as the optimal parameter to control the integrated energy system.

[0009] Preferably, in the third stage, a random position X is generated. p4 The method for (i,:) is as follows:

[0010] X p4 (i,:)=X(i,:)+r1·(X best -|X(i,:)|)-r2·(X worst -|X(i,:)|)+random()×(LO_LOCAL+D×(HI_LOCAL-LO_LOCAL))

[0011] Among them, X(i,:) is the original position of the individual in the current population; r1 and r2 are the weight coefficients of the best individual position and the worst individual position; X best and X worst are the optimal individual position and the worst individual position in the current population respectively; random() is a random array; LO_LOCAL and HI_LOCAL are the lower bound and upper bound of the local search respectively; D is a randomly generated coefficient.

[0012] Preferably, in the first stage, a plurality of individuals are randomly selected from the current population, the selected individuals are divided into two parts, and the positions of the two parts of individuals are updated respectively; the updated positions X of the two parts of individuals are P1 (i,:) and X P2 (i,:) are represented as:

[0013] X P1 (i,:)=X(i,:)+A×(X best -I1×X(i,:))

[0014] X P2(i,:)=X(i,:)+B×(X best -I2×MeanGroup)

[0015] Among them, X(i,:) is the original position of the i-th individual in the current population; X best is the optimal individual position; MeanGroup is the average position of randomly selected individuals; A and B are the weights of the two parts of individuals respectively; I1 and I2 are the coefficients of the original position and the average position respectively;

[0016] Based on the updated position X P1 (i,:) and update position X P2 (i,:) updates the original position of individuals in the current population.

[0017] As a preference, when obtaining the updated position X P1 (i,:) and X P2 (i,:) is followed by a limit on the update position. The limit method is as follows:

[0018] To update position X P1 (i,:) and X P2 (i,:) as the initial update position ; Limit the initial update position to get the final update position , which is expressed as follows:

[0019]

[0020] Among them, u and l are the upper and lower bounds of the search space respectively.

[0021] As an advantage, in the second stage, the positions of the individuals not selected in the first stage are updated, and the updated positions X of the individuals in the second stage are updated. P3 (i,:) is represented as:

[0022]

[0023] Where RL(i,:) is the step length with Rhine flight; p is the position of the threat; distance is the distance between the individual in the current population and the threat; b, c, d are random coefficients; l is the lower bound of the search space;

[0024] Based on the updated position X P3 (i,:) updates the original position of individuals in the current population.

[0025] Preferably, in the second stage, the method for updating the original positions of individuals in the current population is as follows:

[0026] The threat actor's position p is:

[0027] p=lv+r·(uv-lv)

[0028] Among them, v is a parameter related to the step size; r is a random coefficient; u and l are the upper and lower bounds of the search space respectively;

[0029] Get the original position X(i,:) and updated position X of the position update individual respectively P3 (i,:) and the fitness corresponding to the threat position p, and the position with the smallest fitness among the three positions is selected as the original position of the individual in the current population.

[0030] Preferably, the constructed comprehensive energy system model includes an electric power system model, a natural gas system model and a thermal system model; the electric power system model is constructed using a DC power flow model.

[0031] Preferably, the objective function is constructed by taking the minimum cost of the integrated energy system as the optimization target of the integrated energy model to construct the objective function; the objective function includes the generator operating cost, the cogeneration unit operating cost, the natural gas cost and the investment cost.

[0032] Preferably, the Latin hypercube sampling method is used to initialize the original positions of individuals in the population.

[0033] In a second aspect, the present invention provides an integrated energy optimization control system based on multi-objective optimization, which is used to execute the above-mentioned integrated energy optimization control method; the integrated energy optimization control system includes a power system, a natural gas system and a thermal system; the power system is used to convert various energy sources into electrical energy and transmit the electricity to various loads; the natural gas system is used to transmit natural gas from the gas source to the power system and the thermal system; the thermal system includes a gas turbine, a cogeneration unit and a power-to-gas device; the gas turbine is used to burn natural gas to supply power to the power system; the cogeneration unit is used to utilize the waste heat of the gas turbine to generate electricity and provide heat; the power-to-gas device is used to convert excess electricity into natural gas for storage.

[0034] The present invention has the following beneficial effects:

[0035] 1. The present invention combines the optimal individual position and the worst individual position to update the original position of the individual in the position update of multi-objective optimization. By adjusting the position toward the optimal individual position and away from the worst individual position in both directions, the global exploration and local development capabilities are automatically balanced, thereby improving the direction diversity. At the same time, the present invention introduces random coefficients for the optimal individual position and the worst individual position, so that the search direction changes dynamically, reducing the probability of falling into the local optimum.

[0036] 2. The present invention uses a reflective boundary processing method to process the updated position, solving the boundary accumulation problem caused by the existing truncated boundary processing method, enhancing the exploration capability near the boundary, and more balancing the global and local search, making it less likely to fall into the local optimum. At the same time, the present invention shows excellent performance in dealing with nonlinear, multi-objective and dynamically changing optimization problems, demonstrating its great application potential in the multi-objective optimization of integrated energy systems. It not only provides a new technical path for the optimization problems of integrated energy systems, but also provides theoretical support and practical reference for energy conservation, emission reduction and green development in the energy field. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 Schematic diagram of the integrated energy optimization control system in the present invention.

[0038] Figure 2 Schematic diagram of the natural gas system model constructed in the present invention.

[0039] Figure 3 This is a flow chart for optimizing the parameters of the integrated energy system model in the present invention.

[0040] Figure 4 This is a schematic diagram of the nodes of the integrated energy system in the present invention.

[0041] Figure 5 This is a three-dimensional diagram of pipeline natural gas flow in the present invention.

[0042] Figure 6 This is a three-dimensional diagram of the air pressure distribution in the present invention.

[0043] Figure 7 This is the power output diagram of the power-to-gas device in the present invention.

[0044] Figure 8 This is a three-dimensional diagram of the power system branch power in the present invention.

[0045] Figure 9 This is the output diagram of the wind turbine generator set in the present invention.

[0046] Figure 10 This is the gas consumption diagram of the natural gas generator in the present invention.

[0047] Figure 11 This is the power output diagram of the thermal power unit in the present invention.

[0048] Figure 12 This is the natural gas source output diagram of the present invention. DETAILED DESCRIPTION

[0049] The present invention will be further described below with reference to the accompanying drawings.

[0050] like Figure 1Figure 1 shows an integrated energy optimization control method based on multi-objective optimization. The integrated energy optimization control system includes a power system, a natural gas system, and a thermal system. The power system converts various energy sources into electricity and delivers this electricity to various loads. The natural gas system transports natural gas from the source to the power and thermal systems. The thermal system includes a gas turbine, a cogeneration unit, and a power-to-gas device. The gas turbine burns natural gas to power the power system; the cogeneration unit utilizes waste heat from the gas turbine to generate electricity and heat; and the power-to-gas device converts excess electricity into natural gas for storage.

[0051] The comprehensive energy optimization control method comprises the following steps:

[0052] Step 1: Build a comprehensive energy system model

[0053] The integrated energy system model includes an electric power system model, a natural gas system model, and a thermal system model. The electric power system model is constructed using a DC power flow model. The DC power flow model is a simplified model in the steady-state analysis of the electric power system, and is mainly used to calculate the active power distribution in the electric power system. This model converts the AC network model of the electric power system into a DC network model, ignoring the influence of frequency and phase in the AC system, and expressing all AC quantities in terms of DC quantities. In the DC network model, the voltage and power flow distribution of each node are obtained by solving the DC power flow equation. The DC power flow model assumes that the voltage amplitude of each node in the power system remains unchanged, and ignores factors such as resistance, charging capacitance, and parallel compensation on the line. This model simplifies the complex nonlinear power flow model into a linear model, which is expressed as:

[0054] P=αβ

[0055] Where P is the net active power injection vector of each node; α is the system susceptance matrix; β is the voltage phase angle vector of each node.

[0056] This simplification allows the model to quickly calculate the power flow in the power system while maintaining sufficient accuracy.

[0057] like Figure 2 As shown, the natural gas system model is constructed as follows:

[0058]

[0059] in, is the flow rate of pipes a and b at time t; s is a constant related to pipe temperature, length, inner diameter, and compressibility; ρ a,t and ρ b,t are the port pressures of pipes a and b at time t respectively.

[0060] Natural gas pipelines are responsible for transporting natural gas from the gas source to various consumption points. The diameter, temperature, pressure and other factors of the pipeline will affect the flow of natural gas. The Weymouth steady-state model is used to describe the flow of natural gas pipelines. That is, the flow is only related to the pressure at both ends of the pipeline, and flows from the node with high pressure to the node with low pressure. The natural gas flow rate L flowing through the pipeline is X-Y The expression is:

[0061]

[0062] Among them, ε X-Y is the pipeline transmission parameter, i.e. temperature, diameter and other factors; P X 、P Y Respectively represent the air pressure of nodes X and Y; sgn(P X ,P Y ) is a direction parameter, indicating the direction of natural gas flow in the pipeline. When it is equal to 1, it means that natural gas flows from node X to node Y. When it is equal to -1, it means that natural gas flows from node Y to node X.

[0063] Because natural gas is affected by its own materials and external factors during transportation, it experiences pressure drops. To maintain the pressure of natural gas in the pipeline, a compressor is required to boost the pressure. Since compressors consume relatively little energy (electricity or natural gas), only the node pressure relationship at both ends of the compressor is retained, ignoring the energy consumed by the compressor. This is expressed as follows:

[0064] P j =ρ c P n

[0065] Among them, P j and P n are the pressures at the compressor outlet and inlet, respectively; ρ c is the compression ratio of the compressor. Since the consumption characteristics of the compressor are not considered, the compression ratio ρ c is a constant.

[0066] The pipeline model in the natural gas system is a mathematical model used to describe and analyze the flow of natural gas in the natural gas system. It is crucial for the design, operation, and optimization of the natural gas transmission system and is expressed as follows:

[0067]

[0068] w xt,t +w yx,t =0

[0069]

[0070]

[0071]

[0072] w xy,min ≤w xy,t ≤w xy,max

[0073]

[0074] Among them, w x,t and w y,t are the injected gas pressures at node x and node y at time t; w xy,t is the gas flow from pipeline node x to node y; w yk,t is the gas flow from pipeline node y to node k; Z(y) is the pipeline network with node y as the end; v(y) is the pipeline network with node y as the beginning; w yx,t is the gas flow rate transmitted from pipeline node y to node x; Inject air flow into the air source point; is the gas consumption of the gas turbine at node y; is the gas load of node y; C xy is the pipeline xy constant, which is related to the pipeline length, pipeline diameter, operating temperature and the pressure difference between nodes; and are the air pressures at nodes x and y, respectively; and w is the upper and lower limits of node pressure; xy,min and w xy,max The upper and lower limits of pipeline transmission airflow; is the gas shedding load at node y.

[0075] The thermal system model includes a heat source model and a P2G (power-to-gas) device model. The heat source model includes a gas turbine model and a combined heat and power (CHP) unit model.

[0076] The gas turbine model is expressed as:

[0077]

[0078] in, is the gas flow consumed by the i-th gas turbine at time t; α i , β i and γ i is the consumption coefficient; is the active power generated by the i-th gas turbine at time t.

[0079] The cost of natural gas fuel consumed by the gas turbine C fuel for:

[0080] C fuel =C ng ·H gas

[0081] Among them, C ng is the natural gas price; H gas is the natural gas consumption of the gas turbine, which is expressed as:

[0082]

[0083] Among them, Q LHV The low calorific value of natural gas is 9.7kwh / m 3 ; P(t) is the power output of the gas turbine in time t; η(t) is the efficiency of the gas turbine in time t; Δt is the time interval.

[0084] The constructed cogeneration unit model is as follows:

[0085] φ CHP (t)=2.58*P CHP (t)

[0086] Among them, φ CHP (t) is the thermal power generated by the waste heat boiler of the CHP unit; P CHP (t) is the electric power output of the CHP unit.

[0087] The constructed P2G model is as follows:

[0088]

[0089] in, is the natural gas flow converted by the P2G device; P P2G is all the work consumed in the power-to-gas process; η P2G is the conversion efficiency; GHV is the calorific value of natural gas.

[0090] In some embodiments, the heat source is connected to the heat load through a heating network. Since a large amount of heat medium is stored in the heating network and the heat medium flow rate is slow, and the heat source and the heat load are far apart, the water supply pipe in the heating network can be regarded as a large heat storage device with heat loss and heat delay characteristics. The heat loss of the water supply pipe is mainly caused by the temperature difference between the heat medium in the pipe and the medium outside the pipe, as well as the pipe diameter and insulation degree, and the heat loss in the short term can be regarded as caused by the temperature change of the heat medium in the pipe. Assuming that the heating pipe is laid in the soil, the heat loss caused by the pipe diameter and insulation degree can be ignored. The heat loss of the water supply pipe is expressed as:

[0091]

[0092] in, is the heat loss of pipe y at time t, MW; C W is the heat capacity of water, which is 4200 J / (kg·℃); is the mass flow rate of water in pipe y, kg / s; is the water inlet temperature of the water supply pipe; is the outlet water temperature of the water supply pipe;

[0093] The relationship between the inlet and outlet water temperatures of the water supply pipe is:

[0094]

[0095] Among them, T SOIL is the soil temperature; m P is the mass flow rate of water, kg / s; θ P is the pipe length; p is the pipe heat transfer coefficient.

[0096] In other embodiments, other existing methods may be used to transfer heat between the heat source and the heat load.

[0097] Step 2: Taking the minimum cost of the integrated energy system as the optimization goal of the integrated energy model, the objective function C of the integrated energy model is constructed as follows:

[0098] C=365×(C1+C2+C3)+C inv

[0099] Among them, C1 is the generator operating cost; C2 is the CHP unit operating cost; C3 is the natural gas cost; C inv is the investment cost, that is, the purchase cost of the equipment.

[0100] The generator operating cost C1 is expressed as follows:

[0101]

[0102] Among them, a x 、b x and c x These are all cost factors of the generator; is the active power output of generator x at time t.

[0103] Since the CHP unit operating cost C2 is mainly fuel cost, it is expressed as follows:

[0104]

[0105] Among them, ε CHP is the fuel cost coefficient of the CHP unit; θ P and θ HRespectively represent the fuel consumed by the CHP unit to generate unit electrical power and unit thermal power; and They represent the electric power and thermal power generated by CHP unit x at time t respectively.

[0106] The natural gas cost C3 is expressed as follows:

[0107] C3=γ x Q x,t

[0108] Among them, γ x is the natural gas cost coefficient; Q x,t is the output value of the natural gas source at time t.

[0109] Step 3: To ensure the safe and stable operation of the integrated energy system, it is necessary to construct the constraints of the integrated energy system model, which are as follows:

[0110] (1) Generator output constraints:

[0111]

[0112] in, are the upper and lower limits of the output of generator i; PG i is the output of generator i.

[0113] (2) Generator unit climbing constraints:

[0114] P G,i,t -P G,i,t-1 ≤X H,i

[0115] P G,i,t-1 -P G,i,t ≤X L,i

[0116] Among them, P G,i,t and P G,i,t-1 They represent the output of generator set i at time t and time t-1 respectively; X H,i and X L,i are the upper and lower climbing limits of generator set i respectively.

[0117] (3) Power system branch flow constraints

[0118] The power flowing through each current branch in the model cannot exceed the maximum and minimum values of the power allowed to be transmitted by the branch, which is expressed as follows:

[0119]

[0120] in, Indicates the maximum power allowed to flow through branch r; is the power flowing through branch r.

[0121] (4) Node power balance constraints

[0122] At the system level, the power balance constraints of all nodes are aggregated to form the power balance equation for the entire system. For each node, its power balance constraint can be expressed as:

[0123]

[0124] in, represents the set of lines connected to node i; represents the set of generators on node i; P ij represents the power flowing from node i to node j; represents the load of node i; P g,i Represents the total output of all generators at node i.

[0125] (5) Node pressure constraint

[0126] To ensure the safe and stable operation of the natural gas system, the gas pressure at each node must be within a reasonable operating range, which can be expressed as:

[0127]

[0128] in, Respectively represent the minimum and maximum values of the air pressure at node X; P X Indicates the air pressure at node X.

[0129] (6) Thermal system balance constraints

[0130] In the equilibrium state, the energy conversion and transfer within the system reaches a stable state, which conforms to the first law of thermodynamics. At the same time, it avoids excessive or insufficient heat production, ensuring the stability of the system, which can be expressed as:

[0131] φ CHP (t)+φ EB (t) = φ loss (t)+φ load (t)

[0132] Among them, φ CHP (t) is the heat energy generated by the CHP unit at time t, φ EB (t) is the heat energy generated by the electric boiler at time t, φ loss (t) is the heat energy lost during the heat production process, φ load (t) is the effective thermal energy.

[0133] Step 4: Figure 3As shown in Figure 2, the specific process of solving the multi-objective optimization problem is as follows:

[0134] 4-1. Population initialization

[0135] The multi-objective integrated energy system model parameters are used as the original positions of the individuals to construct a population. The population size N, decision dimension m, total number of iterations J, upper bound u of the search space, and lower bound l of the search space are initialized. The Latin Hypercube Sampling (LHS) method is used to initialize each individual position X0(i,:) in the population. Latin Hypercube sampling is based on the Latin Hypercube sampling function, which returns a one-dimensional matrix. The sampling distribution of Latin Hypercube sampling is more uniform than random sampling. Therefore, the algorithm using Latin Hypercube sampling for population initialization helps avoid the population from getting stuck during the initialization phase. The individual position X0(i,:) is represented as follows:

[0136] X0(i,:)=(ul)×lhs(1,m)+l

[0137] Wherein, lhs(1,m) represents the Latin hypercube sampling function.

[0138] The objective function of the comprehensive energy model is used as the fitness function; the fitness of each individual is obtained according to the original position of the individual in the population, which is used for position update and strategy selection in subsequent stages.

[0139] 4-2. Phase 1

[0140] Randomly select some individuals (N / 2) from the current population and randomly divide the selected individuals into two parts, and update their positions separately, which is expressed as follows:

[0141] X P1 (i,:)=X(i,:)+A×(X best -I1×X(i,:))

[0142] X P2 (i,:)=X(i,:)+B×(X best -I2×MeanGroup)

[0143] Among them, X P1 (i,:) and X P2 (i,:) are the updated positions of the two parts of individuals; X(i,:) is the original position of the i-th individual in the current population; X best is the position of the optimal individual; MeanGroup is the average position of some randomly selected individuals; A and B are the weights of the two parts of individuals, and their values are [0,1]; I1 and I2 are the coefficients of the original position and the average position, respectively, which are random integers.

[0144] The update position is processed by boundary processing, and the specific process is as follows:

[0145] To update position X P1 (i,:) and X P2 (i,:) as the initial update position Based on the search space upper bound ub and search space lower bound lb, the initial update position is restricted to obtain the final update position It is expressed as follows:

[0146]

[0147] The fitness corresponding to the original position and the final updated position of the position updated individual are obtained respectively. If the fitness corresponding to the original position is greater than the fitness corresponding to the final updated position, the final updated position is used as the original position of the individual in the current population.

[0148] 4-3. Phase 2

[0149] The positions of another part of individuals in the current population (individuals that have not undergone the position update in stage 2) are updated, which is expressed as:

[0150]

[0151] Among them, X P3 (i,:) is the updated position of the i-th individual; RL(i,:) is the step length with Rhine flight; p is the position of the threat; distance is the distance between the individual and the threat in the current population; b, c, d are random coefficients.

[0152] The threat actor's position p is set to:

[0153] p=lv+r·(uv-lv)

[0154] Among them, v is a parameter related to the step size RL(i,:); r is a random coefficient.

[0155] Perform boundary processing on the updated position; obtain the fitness corresponding to the original position X(i,:) of the position update individual and the threat position p, as well as the fitness corresponding to the updated position after boundary processing, and select the position with the smallest fitness among the three positions as the original position of the individual in the current population.

[0156] 4-4. Stage 3

[0157] Generate a random position X based on the original position of the individual in the current population p4 (i,:), which is represented as follows:

[0158] X p4(i,:)=X(i,:)+r1·(X best -|X(i,:)|)-r2·(X worst -|X(i,:)|)+random()×(LO_LOCAL+D×(HI_LOCAL-LO_LOCAL))

[0159] Among them, random() is a random array; LO_LOCAL and HI_LOCAL are the lower and upper bounds of the local search respectively; D is a randomly generated coefficient; r1 and r2 are the random coefficients of the best individual position and the worst individual position, which are random numbers between 0 and 1; X best and X worst are the best and worst individual positions in the current population respectively; |·| is the absolute value symbol, which aims to convert all elements in the original position into non-negative numbers to avoid directional cancellation or interference of positive and negative values when generating random positions, and ensure that calculations are performed only based on numerical values.

[0160] In this embodiment, the best individual position and the worst individual position are the positions corresponding to the minimum fitness and the maximum fitness, respectively.

[0161] Perform boundary processing on the random position; obtain the fitness corresponding to the original position X(i,:) of the individual in the current population and the fitness corresponding to the random position after boundary processing respectively. If the fitness corresponding to the random position after boundary processing is less than the fitness corresponding to the original position X(i,:), then replace the original position of the corresponding individual in the current population with the random position after boundary processing.

[0162] By introducing the best and worst individual positions during the development phase and updating the positions, we can continuously approach the best individual while moving away from the worst individual, thereby continuously improving the quality of the solution. This increases the convergence speed and allows us to quickly find the optimal solution without having to adjust too many parameters.

[0163] 4-5. Repeat the above steps until the iteration termination condition is met, that is, the fitness corresponding to the optimal individual position is less than the preset threshold or the number of iterations is equal to the total number of iterations; set the model parameters of the integrated energy system based on the position corresponding to the optimal individual position to obtain the optimal integrated energy optimization control scheme.

[0164] Step 5: Method Evaluation

[0165] The structure of all nodes included in the integrated energy system is as follows Figure 4As shown in Figure 1, nodes 33 and 37 in the IEEE-39 power system are configured as gas-fired generators, with their natural gas supply coming from nodes 6 and 19, respectively, of the 20-node natural gas system. Meanwhile, the generator at node 30 is configured as a combined heat and power (CHP) unit, which not only generates electricity but also provides heat to the 6-node thermal system. The 20-node natural gas system includes six natural gas supply points and nine demand points. The 6-node thermal system consists of a CHP unit and an electric boiler, whose electricity consumption depends on the power supplied by the CHP unit. To enhance the interconnection between electricity, natural gas, and thermal energy, the CHP unit is designed as a gas-fired unit, with its natural gas consumption supplied by node 3 of the natural gas system. The power system data is identical to that of the matpower39 node, with a total of 10 generators, two of which have been converted to gas-fired generators. The total installed capacity is 6,967 MW, and the total power load is 5,941.5 MW. Table 1 shows the coupling of the relevant nodes in the power system.

[0166] Table 1 Description of important nodes

[0167] Node number Machine Type Gas input source Remark Grid node 33 gas turbine Gas network node 6 gas supply none Grid Node 37 gas turbine Gas network node 19 gas supply none Grid node 30 CHP unit Gas network node 3 gas supply It is the heat source of heating network node 1 The remaining 7 power nodes coal-fired units none none

[0168] Some nodes in the power system can serve as gas sources in the natural gas system and heat sources in the thermal system. The cost parameters of the generator sets are shown in Table 2.

[0169] Table 2 Generator cost parameters

[0170]

[0171]

[0172] Among them, a x 、b x and c x are the cost coefficients of the generator respectively.

[0173] This case includes 6 gas sources and 9 gas loads. 7 conventional gas loads and 2 gas turbine generator loads. The total load is 2.4608Mm 3 The cost of gas sources 1 to 3 is $0.085 / m 3 The cost of gas sources 4 to 6 is $0.062 / m 3 The upper and lower limits of the flow rate for each gas source are shown in Table 3.

[0174] Table 3 Basic parameters of natural gas source

[0175] Gas source number <![CDATA[Flow lower limit / Mm 3 > <![CDATA[Flow upper limit / Mm 3 > 1 0.90 1.7391 2 0 1.26 3 0 0.72 4 1.00 2.3018 5 0 0.27 6 0 1.44

[0176] The natural gas data is for a 20-node natural gas system. Table 4 shows the pipeline data for the 20-node natural gas system, and Table 5 shows the node data for the 20-node natural gas system.

[0177] Table 4 20-node natural gas system pipeline data

[0178]

[0179]

[0180] Among them, c k-n is the flow from node k to node n in the natural gas pipeline.

[0181] Table 520 Node Natural Gas System Node Data

[0182]

[0183] The thermal system consists of one CHP unit, one electric boiler, and three heat loads. The total load is 50 MW, with the electric boiler's power-to-heat ratio of 0.8 and an output limit of 30 MW. The basic parameters of the CHP unit are shown in Table 6.

[0184] Table 6 CHP unit parameters

[0185]

[0186] In Table 6, ε CHP is the fuel cost coefficient of the CHP unit, θ P ,θ H They represent the fuel consumed by the CHP unit to generate unit electrical power and unit thermal power respectively.

[0187] Figure 4 The ambient temperatures of the six nodes in the thermal system are shown in Table 7.

[0188] Table 7 Ambient temperature of 6 nodes in the thermal system

[0189]

[0190]

[0191] This case was solved using both the Particle Swarm Optimization (PSO) and the Pigeon Swarm Optimization (PSO) algorithms, and the results were compared with those obtained using the present invention. The number of iterations was set as follows: 30 global search iterations and 5 local search iterations for the PSO algorithm; 30 iterations for both the PSO algorithm and the present invention. The common parameter for all three algorithms was the initial population size of 6. Table 8 compares the cost optimization results for the three algorithms in this case.

[0192] Table 8 Comparison of cost optimization results of three algorithms

[0193]

[0194] It can be seen from the cost optimization results in Table 8 that, compared with the pigeon swarm optimization algorithm and the particle swarm optimization algorithm, the present invention has greatly reduced the wind turbine operating cost after multiple iterative solutions. Although there has been some increase in the cost of other parts, the overall cost reduction goal has been achieved, with the overall cost reduced by about 1.5%.

[0195] The three-dimensional graph of pipeline natural gas flow of the present invention is as follows Figure 5 As shown in the figure, the flow rates of branches 1 to 5 remain between 0 and 0.5 m³ / h in the morning, then gradually increase to between 0.5 and 1 m³ / h. At branch 10, the pipeline natural gas flow rate is the highest of all branches, averaging around 1.5 m³ / h. The curve shows significant fluctuations throughout the day, indicating that gas flow rates vary significantly across time periods. In some periods, flow rates reach peaks, while in other periods, flow rates may drop to near zero or even negative values, indicating a decrease or reverse flow. The primary factors influencing pipeline natural gas flow are various natural gas pipeline parameters.

[0196] The three-dimensional diagram of air pressure distribution is as follows Figure 6 As shown, air pressure data may often reflect key operational characteristics or control mechanisms in the model. Periodic pressure fluctuations may correspond to intermittent power input from wind turbines, indicating that the gas grid is actively absorbing excess power through power-to-gas equipment. A fluctuation phase difference of less than 15% can be considered effective peak shaving and valley filling. This figure shows that the present invention exhibits a wide pressure range of 20-70 Pa, achieving dynamic energy balance through periodic pressure oscillations (period 2.5 hours ± 5%). Its pressure variation curve shows an 82.3% phase match with the P2G system load demand.

[0197] In the electricity-gas-heat integrated energy system, the output of the P2G (Power-to-Gas) unit is the core indicator reflecting the dynamic characteristics of multi-energy coupling. The output of the P2G unit of the present invention is as follows: Figure 7 As shown, from Figure 7 As can be seen, the present invention exhibits large fluctuations, with a high average power output and a peak power of approximately 33MW. The increased output of P2G units provides technical support for large-scale renewable energy generation, indirectly reducing the use of fossil fuels and other mineral fuels for power generation. This technology strengthens the connection between power and natural gas systems, facilitates the coordinated and optimized operation of multiple energy systems, and improves economic and social benefits.

[0198] The 24-hour power system branch power situation of the present invention is as follows Figure 8 As shown. Figure 8It can be seen that the power of each power system branch calculated by the present invention fluctuates roughly in the range of -0.3MVA to 0.3MVA. The power of this strategy shows obvious volatility within a day, especially the power of branches 1 to 10, which gradually increase during the day, but the overall trend of change is relatively stable. In addition, the power of all branches shows high stability without drastic fluctuations. The power distribution of the present invention is relatively compact, indicating that the present invention shows high stability when processing branch power. The distribution of the maximum value area is relatively uniform, indicating that the performance of the algorithm under different conditions is relatively consistent. The symmetry is relatively high, indicating that the power distribution of the algorithm under different conditions is relatively balanced.

[0199] The output of the wind turbine generator system of the present invention is as follows Figure 9 As shown, from Figure 9 The power output of the present invention varies moderately, from -2 to 12 MW, with a relatively balanced overall output. The power curves of the different units are relatively consistent, demonstrating the present invention's excellent performance in multi-unit collaborative operation. For most of the time period, power output remains high with minimal fluctuations, demonstrating high efficiency and stability. This present invention achieves a good balance between power output and stability, making it suitable for most application scenarios.

[0200] The gas consumption of the natural gas generator of the present invention is as follows Figure 10 As shown, from Figure 10 It can be seen that the gas consumption of the unit 1 of the present invention is stable at 0.006Mm 3 Although the gas consumption of unit 2 fluctuated significantly, it stabilized at 0.0205Mm after 8 hours. 3 It can be seen that the present invention is more inclined to use natural gas for power generation. At the same time, natural gas is a clean energy, which means that the present invention pays more attention to using clean energy to achieve the effect of protecting the environment.

[0201] The output of the thermal power unit of the present invention is as follows Figure 11 As shown. Figure 11 We can clearly see that the output power of the thermal power unit of the present invention fluctuates between 115 and 160MW before 10 hours, and then stabilizes at 160MW. It can be seen that the present invention accounts for a large proportion in the output of the thermal power unit.

[0202] The natural gas source output of the present invention is as follows Figure 12 As shown, from Figure 12 The natural gas source output of the present invention can be seen in the figure, with an average of about 585MW and a maximum of 640MW. It can be seen that the present invention uses a high proportion of clean energy natural gas.

[0203] The present invention shows significant advantages in stability, robustness and solution quality. Thanks to its powerful computing power and efficient global search performance, the algorithm provides an innovative solution to the multi-objective optimization problem of integrated energy systems. Compared with the pigeon flock optimization algorithm, the present invention shows higher optimization accuracy and successfully reduces the total annual cost of the system by 1.5%. In addition, through actual case analysis, the significant effect of the present invention in improving energy utilization and reducing carbon emissions is further demonstrated: energy utilization has increased by about 26.3% and carbon emissions have decreased by about 43%. This achievement is of great significance in the context of global energy transformation and climate change response. It not only improves the economic benefits of the system, but also makes a positive contribution to environmental protection.

[0204] In summary, the present invention demonstrates excellent performance in dealing with nonlinear, multi-objective and dynamically changing optimization problems, showing its great application potential in multi-objective optimization of integrated energy systems. It not only provides a new technical path for the optimization problems of integrated energy systems, but also provides theoretical support and practical reference for energy conservation, emission reduction and green development in the energy field.

Claims

1. A comprehensive energy optimization control method based on multi-objective optimization, characterized in that: include: Construct an integrated energy system model and set the objective function and constraints of the integrated energy system; The population is constructed with the parameters of the integrated energy system model as the original positions of the individuals, and the objective function of the integrated energy system model is used as the fitness function. The parameters in the integrated energy system model are optimized by iteratively updating the original positions of the individuals. In one iteration, the individual positions are updated in the first, second, and third stages. The first stage obtains an updated position based on the original position of the individual, and updates the individual position according to the updated position; the second stage constructs the position of the threat, obtains an updated position based on the position of the threat, and updates the individual position according to the updated position; the third stage generates a random position by introducing the best individual position and the worst individual position in the current population, and updates the position based on the random position; After reaching the iteration termination condition, the individual position with the minimum fitness is used as the optimal parameter to control the integrated energy system.

2. The comprehensive energy optimization control method based on multi-objective optimization according to claim 1, characterized in that: In the third stage, a random position X is generated p4 The method for (i,:) is as follows: X p4 (i,:)=X(i,:)+r1·(X best -|X(i,:)|)-r2·(X worst -|X(i,:)|)+random()×(LO_LOCAL+D×(HI_LOCAL-LO_LOCAL)) Among them, X(i,:) is the original position of the individual in the current population; r1 and r2 are the random coefficients of the best individual position and the worst individual position; X best and X worst are the optimal individual position and the worst individual position in the current population respectively; random() is a random array; LO_LOCAL and HI_LOCAL are the lower bound and upper bound of the local search respectively; D is a randomly generated coefficient.

3. The comprehensive energy optimization control method based on multi-objective optimization according to claim 1, characterized in that: After obtaining the updated position or random position, the updated position or random position is subjected to boundary processing, which is expressed as follows: in, Indicates the position after boundary processing; Indicates the position before boundary processing; u and l are the upper and lower bounds of the search space respectively.

4. The comprehensive energy optimization control method based on multi-objective optimization according to claim 1, characterized in that: In the first stage, multiple individuals are randomly selected from the current population, the selected individuals are divided into two parts, and the positions of the two parts of individuals are updated respectively; the updated positions X of the two parts of individuals are P1 (i,:) and X P2 (i,:) are represented as: X P1 (i,:)=X(i,:)+A×(X best -I1×X(i,:)) X P2 (i,:)=X(i,:)+B×(X best -I2×MeanGroup) Among them, X(i,:) is the original position of the i-th individual in the current population; X best is the optimal individual position; MeanGroup is the average position of randomly selected individuals; A and B are the weights of the two parts of individuals respectively; I1 and I2 are the coefficients of the original position and the average position respectively; Based on the updated position X P1 (i,:) and update position X P2 (i,:) updates the original position of individuals in the current population.

5. The comprehensive energy optimization control method based on multi-objective optimization according to claim 4 is characterized in that: In the second stage, the positions of the individuals not selected in the first stage are updated. The updated positions X of the individuals in the second stage are P3 (i,:) is represented as: Where RL(i,:) is the step length with Rhine flight; p is the position of the threat; distance is the distance between the individual in the current population and the threat; b, c, d are random coefficients; l is the lower bound of the search space; Based on the updated position X P3 (i,:) updates the original position of individuals in the current population.

6. The comprehensive energy optimization control method based on multi-objective optimization according to claim 5, characterized in that: In the second stage, the method for updating the original position of individuals in the current population is as follows: The threat actor's position p is: p=lv+r·(uv-lv) Among them, v is a parameter related to the step size; r is a random coefficient; u and l are the upper and lower bounds of the search space respectively; Obtain the fitness corresponding to the original position X(i,:) of the position update individual, the updated position after boundary processing, and the threat position p respectively, and select the position with the smallest fitness among the three positions as the original position of the individual in the current population.

7. The comprehensive energy optimization control method based on multi-objective optimization according to claim 1, characterized in that: The constructed comprehensive energy system model includes power system model, natural gas system model and thermal system model; the power system model is constructed using the DC power flow model.

8. The comprehensive energy optimization control method based on multi-objective optimization according to claim 1, characterized in that: The objective function is constructed by taking the minimum cost of the integrated energy system as the optimization target of the integrated energy model to construct the objective function; the objective function includes the generator operating cost, the cogeneration unit operating cost, the natural gas cost and the investment cost.

9. The comprehensive energy optimization control method based on multi-objective optimization according to claim 1, characterized in that: The Latin hypercube sampling method is used to initialize the original positions of individuals in the population.

10. A comprehensive energy optimization control system based on multi-objective optimization, characterized by: Used to execute the integrated energy optimization control method based on multi-objective optimization as described in claim 1; the integrated energy optimization control system includes an electric power system, a natural gas system and a thermal system; the electric power system is used to convert various energy sources into electrical energy and transmit the electricity to various loads; the natural gas system is used to transmit natural gas from the gas source to the electric power system and the thermal system; the thermal system includes a gas turbine, a cogeneration unit and a power-to-gas device; the gas turbine is used to burn natural gas to supply power to the electric power system; the cogeneration unit is used to utilize the waste heat of the gas turbine to generate electricity and provide heat; the power-to-gas device is used to convert excess electricity into natural gas for storage.

Images