Scheduling optimization method and device for comprehensive energy system of agricultural park
By constructing a two-layer model and iterative optimization algorithm for the integrated energy system of agricultural parks, the problems of neglecting carbon emissions and user comfort in traditional scheduling strategies are solved, achieving low-carbon and economical operation and system stability, and optimizing energy purchase costs.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA POWER ENGINEERING CONSULTING GROUP CORPORATION
- Filing Date
- 2025-12-31
- Publication Date
- 2026-04-21
AI Technical Summary
Traditional integrated energy system scheduling strategies in agricultural parks neglect carbon emission costs and user comfort, making it difficult to coordinate the competing interests between the energy and user ends, leading to unstable system operation.
A two-layer model is constructed, encompassing both the energy and user ends. By establishing equipment operation and load models, and considering factors such as the uncertainty of renewable energy, straw procurement and transportation costs, and carbon trading costs, an iterative optimization algorithm is employed to determine the optimal scheduling scheme, thereby achieving a game balance between the energy and user ends.
It has enabled the low-carbon and economical operation of the integrated energy system in agricultural parks, coordinated the interests of the energy end and the user end, improved system stability and user comfort, and reduced energy purchase costs.
Smart Images
Figure CN121903286A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power system dispatching and operation technology, and in particular to a dispatching optimization method and apparatus for an integrated energy system in an agricultural park. Background Technology
[0002] The development of power grids in remote areas lags behind, and power shortages are a key factor hindering the development of their industrial parks. Compared to cities, rural areas have distinct regional characteristics and abundant renewable energy sources, such as wind, solar, and biomass energy. Consuming local renewable energy can save on the construction costs of grid-related facilities, enabling rural industrial parks to achieve energy self-sufficiency. Simultaneously, developing straw biomass energy supply helps to fundamentally reduce carbon emissions. Therefore, research on low-carbon dispatch strategies and devices for off-grid integrated energy systems in agricultural industrial parks is of great significance for achieving energy self-sufficiency in rural industrial parks and the energy utilization of straw biomass.
[0003] In related technologies, the scheduling of integrated energy systems in agricultural parks involves multiple factors. Traditional scheduling strategies ignore carbon emission costs, and research lacks consideration for user comfort (such as thermal comfort) and energy consumption experience, making it difficult to coordinate the competition of interests between the energy side, composed of IES operators (IESOs), and the user side, composed of load aggregators (LAs), thus making it difficult to maintain the stable operation of the system.
[0004] Therefore, there is an urgent need for a scheduling optimization method and device for the integrated energy system of agricultural parks to solve the above-mentioned technical problems. Summary of the Invention
[0005] This invention provides a scheduling optimization method and apparatus for an integrated energy system in an agricultural park, which can achieve low-carbon and economical operation of the integrated energy system in the agricultural park. The technical solution is as follows: On the one hand, a scheduling optimization method for an integrated energy system in an agricultural park is provided, the method comprising: Based on the operational data of the integrated energy system in the agricultural park, establish an equipment operation model for the energy side and a load model for the user side; Based on the system equipment operation model, a model for maximizing upper-level operational revenue, including straw procurement and transportation costs, is established. Based on the load model and energy sales price, establish a model for minimizing the energy purchase cost at the user end. Based on the constraints of energy sales price and system balance, the upper-level operating revenue maximization model and the lower-level energy purchase cost minimization model are iteratively optimized to obtain the optimal scheduling scheme of the agricultural park's integrated energy system.
[0006] On the other hand, a scheduling and optimization device for an integrated energy system in an agricultural park is provided, the device comprising: The first modeling module is used to establish equipment operation models on the energy side and load models on the user side based on the operation data of the integrated energy system of the agricultural park. The second modeling module is used to establish an upper-level operational revenue maximization model that includes straw procurement and transportation costs on the energy side, based on the system equipment operation model. The third modeling module is used to establish a model for minimizing the energy purchase cost at the user end based on the load model and energy sales price. The optimization module is used to iteratively optimize the upper-level operating revenue maximization model and the lower-level energy purchase cost minimization model based on energy sales price constraints and system balance constraints, and calculate the optimal scheduling scheme of the agricultural park's integrated energy system.
[0007] On the other hand, a computer device is provided, the computer device including a memory and a processor, the memory for storing computer programs, and the processor for executing the computer programs stored in the memory to implement the steps of the scheduling optimization method for the integrated energy system of the agricultural park described above.
[0008] On the other hand, a computer-readable storage medium is provided, wherein a computer program is stored therein, and when the computer program is executed by a processor, it implements the steps of the scheduling optimization method for the integrated energy system of the agricultural park described above.
[0009] On the other hand, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the scheduling optimization method for the integrated energy system of the agricultural park described above.
[0010] The technical solution provided by this invention can bring at least the following beneficial effects: First, based on the operation data of the integrated energy system of the agricultural park, an equipment operation model for the energy side and a load model for the user side are established; then, considering the uncertainty of renewable energy, straw purchase and transportation costs, carbon trading costs, and comprehensive demand response, a two-level game model is established, namely, an upper-level operating revenue maximization model and a lower-level energy purchase cost minimization model; finally, the two-level game model is optimized through a preset iterative optimization algorithm to obtain the game equilibrium point between the energy side and the user side, thereby determining the optimal energy purchase price, operating revenue, and energy purchase cost. Attached Figure Description
[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0012] Figure 1 This is a flowchart of a scheduling optimization method for an integrated energy system in an agricultural park, provided by an embodiment of the present invention. Figure 2 This is a structural diagram of an off-grid integrated energy system provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of a leader-follower game architecture framework provided in an embodiment of the present invention; Figure 4 (a) and Figure 4 (b) are schematic diagrams illustrating the clustering effects of wind power generation and photovoltaic power generation provided in an embodiment of the present invention; Figure 5 This is a schematic diagram showing the comparison results of indoor temperature considering PMV and ignoring PMV according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the convergence process of a leader-follower game provided in an embodiment of the present invention; Figure 7 (a) and Figure 7 (b) are schematic diagrams showing the results of energy purchase price and power optimization and thermal optimization respectively provided by an embodiment of the present invention; Figure 8 This is a schematic diagram of the transferable load curve within a scheduling period provided by an embodiment of the present invention; Figure 9 (a) to Figure 9 (c) are schematic diagrams of power supply and demand balance, heating supply and demand balance and natural gas supply and demand balance provided in an embodiment of the present invention. Detailed Implementation
[0013] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.
[0014] As mentioned earlier, traditional scheduling strategies lack consideration for user comfort and energy consumption experience, making it difficult to coordinate the competition of interests between the energy side and the user side, and thus making it difficult to maintain the stable operation of the system.
[0015] Based on this, the concept of this invention is to find the equilibrium point of the game between the energy end and the user end by constructing a two-layer model, thereby achieving the stable operation of the integrated energy system.
[0016] The following describes the specific implementation of the above concept.
[0017] Please refer to Figure 1 The present invention provides a scheduling optimization method for an integrated energy system in an agricultural park, the method comprising: Step 100: Based on the operation data of the integrated energy system of the agricultural park, establish the equipment operation model on the energy side and the load model on the user side; Step 102: Based on the system equipment operation model, establish an upper-level operation revenue maximization model that includes straw procurement and transportation costs at the energy end; Step 104: Based on the load model and energy sales price, establish a model for minimizing the energy purchase cost at the user end. Step 106: Based on the constraints of energy sales price and system balance, iteratively optimize the upper-level operating revenue maximization model and the lower-level energy purchase cost minimization model to calculate the optimal scheduling scheme of the agricultural park's integrated energy system.
[0018] In this embodiment of the invention, firstly, based on the operational data of the integrated energy system of the agricultural park, an equipment operation model for the energy end and a load model for the user end are established; then, considering the uncertainty of renewable energy, straw purchase and transportation costs, carbon trading costs, and comprehensive demand response, a two-layer game model is established, namely, an upper-layer operational revenue maximization model and a lower-layer energy purchase cost minimization model; finally, the two-layer game model is optimized through a preset iterative optimization algorithm to obtain the game equilibrium point between the energy end and the user end, thereby determining the optimal energy purchase price, operational revenue, and energy purchase cost.
[0019] The following description Figure 1 The execution method for each step is shown.
[0020] First, for step 100, based on the operation data of the integrated energy system of the agricultural park, establish the equipment operation model on the energy side and the load model on the user side.
[0021] To fully utilize agricultural biomass resources, this embodiment, considering straw energy utilization, constructs as follows: Figure 2 The off-grid integrated energy system structure is shown, and the energy flow relationships and leader-follower game patterns among system stakeholders are analyzed.
[0022] Specifically, load aggregators (LAs) include both electrical and thermal loads. IESO possesses a variety of energy supply, energy conversion, and energy storage equipment to meet the diverse energy needs of LAs. Energy supply includes renewable energy generation (RE) and pyrolysis gasification (PG) units. Energy conversion includes heat pumps (HP), combined heat and power (CHP), and power-to-gas (P2G) units. Energy storage includes electrical energy storage (EES), thermal energy storage (HES), and gas storage tanks (GS).
[0023] Figure 2 The energy flow relationships of off-grid IES are shown. Renewable energy generation prioritizes meeting electricity load demand, with surplus electricity used for heat pump heating and P2G methane production, or stored in EES. Cogeneration can provide electricity and heat to LA, with surplus heat stored in HES. P2G includes an electrolyzer (EL), a methane reactor (MR), and a carbon capture unit (CCS). Carbon dioxide produced by CHP can be captured by the CCS and supplied to the MR. P2G and PG can produce natural gas to meet the energy needs of CHP, while GS maintains the natural gas supply and demand balance through natural gas storage and release.
[0024] Off-grid IES (Industrial Engineering Systems) employ a leader-follower game model among multiple stakeholders. First, the LA (Local Area) determines load reduction and shifting based on the IESO's initial energy sales price, considering comfort and energy consumption experience. Then, the IESO optimizes its energy sales price based on the LA's energy demand and adjusts the operating power of various units. It's worth noting that excessively high IESO energy sales prices will reduce LA's energy demand, leading to decreased IESO profits; conversely, excessively low IESO energy sales prices, while increasing LA's energy demand, will also reduce IESO profits. Therefore, the leader-follower game model achieves an equilibrium between IESO profits and LA costs, preventing the aforementioned issues.
[0025] Based on this, and using the operational data of the agricultural park's integrated energy system, an operational model for the energy-side equipment is established, including: Based on the probability distribution characteristics of wind speed and solar intensity, a probabilistic model for renewable energy power generation is established.
[0026] Specifically, to ensure the safe and reliable operation of off-grid IES, probabilistic models for wind and solar power generation considering the uncertainties of renewable energy sources were established. Based on this, a large number of scenarios were first generated using Monte Carlo sampling, and then the scale of the scenarios was reduced using k-means clustering, thereby obtaining... Each scenario and its corresponding probability The formula is shown below: In the formula, This is the power of the fan (WT). This refers to the rated power of the fan. Let be the wind speed at time t. To determine the cut-in wind speed. To cut off the wind speed. This is the rated wind speed. , These are the scale parameter and shape parameter of the Weibull distribution, respectively. It is the power of photovoltaic (PV). This is the rated power of the photovoltaic system. It is the light intensity at time t. This is the rated light intensity. It is the temperature-power conversion coefficient. It is the ambient temperature at time t. , These are the shape parameters of the beta distribution.
[0027] A linear gas production model for a pyrolysis gasification device is established based on the energy conversion efficiency of straw.
[0028] Specifically, pyrolysis gasification (PG) units can convert biomass such as straw into natural gas, providing a stable energy supply for industrial production, as shown in the following formula: In the formula, Let be the gas production of PG at time t. This represents PG's efficiency, taken as 0.72. This represents the amount of straw consumed by PG at time t. The value of straw is expressed as 14 MJ / kg. The calorific value of methane can be taken as 35.87 MJ / m³. 3 . This indicates the upper limit of PG straw consumption.
[0029] Based on the electrothermal conversion efficiency and cogeneration operation data, a linear energy conversion model for heat pumps and cogeneration is established.
[0030] Specifically, HP can convert electricity into heat to meet part of the heat demand of off-grid IES, as shown in the following formula: In the formula, Let t be the heat power of the heat pump at time t. The electrothermal conversion coefficient of the heat pump is taken as 3.5. It is the electrical power of HP at time t. This is the upper limit of HP's thermal power.
[0031] In the formula, This represents the electrical power generated by the combined heat and power (CHP) at time t. This represents the power generation efficiency of the combined heat and power (CHP) system, taken as 0.35. This indicates the amount of natural gas consumed in combined heat and power (CHP) generation. This represents the thermal power of CHP at time t. This represents the thermal efficiency of CHP, taken as 0.45. This indicates the upper limit of CHP's electricity.
[0032] Based on electrochemical principles and carbon capture technology, an energy consumption model for the multi-stage chemical reaction of an electrochemical gas generator is established.
[0033] Specifically, the P2G (Power-to-Gas) unit consists of a CCS (Carbon Dioxide Control System), an EL (Electric Electrochemical Unit), and an MR (Metallurgical Regeneration Unit). The CCS captures carbon dioxide produced by CHP (Carbon Dioxide) and supplies it to the MR. The energy consumption of the CCS includes operating energy consumption and stationary energy consumption, as shown in the following formula: In the formula, That is the electrical power of the P2G. This refers to the energy consumption of the CCS. It is the electrical power of EL. This indicates the energy consumption during operation. It is a fixed energy consumption. This is the energy required to capture each unit of carbon dioxide, taken as 0.269 kW / kg. It represents the amount of carbon dioxide captured in time t. This is the upper limit of P2G power.
[0034] The hydrogen gas produced by the EL reaction with CO2 in the MR to produce methane, as shown in the following formula: In the formula, This represents the amount of CO2 consumed by MR at time t. It is the electrical power of EL at time t. is the CO2 consumption coefficient of MR, taken as 0.14 kg / kW. This is the efficiency of EL, taken as 0.7. This is the methane conversion efficiency, taken as 0.88. It is the natural gas produced by MR at time t. This is the upper limit of the EL's electrical power.
[0035] Based on the law of conservation of energy and the characteristics of charge and discharge efficiency, a dynamic capacity model for an electric / thermal energy storage device is established.
[0036] Specifically, EES helps to mitigate power fluctuations from renewable energy sources and maintain a balance between power supply and demand. The remaining capacity of the EES needs to remain consistent at the beginning and end of the dispatch cycle. The HES model is similar to EES, as shown in the following formula: In the formula, This represents the remaining capacity at time t. , These represent the efficiency of energy storage and release, respectively, with a value of 0.95 in EES and 0.98 in HES. , These represent the power stored and released at time t, respectively. , This indicates the remaining capacity at the beginning and end of each scheduling cycle. , It represents the state of energy storage and release, and consists of 0-1 variables. It is the upper limit of power for energy storage and release. , These are the upper and lower limits of the remaining capacity.
[0037] Based on the operating parameters of gas storage, a seasonal gas storage model for the gas storage tank is established.
[0038] Specifically, because gas storage tanks (GS) can achieve seasonal gas storage, the remaining capacity of GS can differ at the beginning and end of each scheduling cycle. The seasonal gas storage model is shown in the following formula: In the formula, Let be the remaining capacity of GS at time t. , The efficiency of gas storage and release in GS is taken as 0.95. , These represent the rates of gas storage and release at time t, respectively. It is the upper limit of the rate of natural gas storage and release in GS. It is the upper limit of the remaining natural gas capacity in GS.
[0039] In this embodiment of the invention, a load model for the user end is established based on the operational data of the integrated energy system of the agricultural park, including: Based on the load's adjustable characteristics and the pre-set tiered excitation mechanism, a demand response model for flexible electrical loads is established.
[0040] Specifically, flexible electrical loads include transferable loads and reduceable loads, as shown below. To ensure the quality of power supply to the loads, the total amount of transferable and reduceable loads shall not exceed 20% of the pre-demand response electrical load.
[0041] In the formula, This represents the electrical load after the demand response at time t. This represents the electrical load before the demand response at time t. This indicates a transferable load. This is the upper limit of transferable load. This allows for load reduction. It is the maximum load that can be reduced.
[0042] To encourage user participation in demand response, a tiered demand response incentive method that can reduce load is proposed, as shown in the following formula: In the formula, This is the demand response compensation provided by IESO to LA. This is the demand response price factor, set at $33.15 per megawatt-hour. It is the interval length for load reduction.
[0043] Based on the thermal comfort index and heat transfer data determined by users through predicted average voting, a user experience-driven model for flexible heat load is established.
[0044] Specifically, the Predicted Average Votes (PMV) index is used to represent the indoor environmental comfort index to describe user thermal comfort. The calculation formula is shown below: In the formula, The user's comfortable temperature is represented by 33.5℃. It is the indoor temperature at time t. This is the human body's energy metabolism rate, taken as 80 W / m 2 . This is the thermal resistance of the clothing, taken as 0.11 (m). 2 ·℃) / W.
[0045] The seven indicators of PMV correspond to the seven thermal sensations in the human body. This indicates the user's optimal thermal comfort state. It indicates slight warmth, warmth, and heat. This indicates slight coolness, coolness, and coldness. To meet the thermal comfort needs of indoor users in winter, the above formula can be rewritten as an indoor temperature model, as shown in the following formula: In the formula, , This indicates the upper and lower limits of PMV.
[0046] When the indoor temperature is controlled within a certain range, users will not feel a significant temperature difference. This effectively improves the flexibility of heat load control, as shown in the following formula: In the formula, The heat load represents time t. It refers to the heating area. The heat transfer coefficient is taken as 1.037 × 10⁻⁶. 5 J / m·℃. Let t be the outdoor temperature at time t. The heat capacity per unit heating area is taken as 1.63 × 10⁻⁶. 5 J / m·℃.
[0047] When a user's thermal comfort decreases, the IESO needs to provide comfort compensation to the LA, as shown in the following formula: In the formula, The compensation coefficient for user thermal comfort is set at 2.21 $ / MW. for The heat load at time t.
[0048] Then, for step 102, based on the system equipment operation model, a model for maximizing upper-level operational revenue, including straw procurement and transportation costs, is established at the energy end.
[0049] The off-grid IES scheduling model constructed in this embodiment of the invention introduces a straw energy utilization mode, comprehensively considering key factors such as the uncertainty of renewable energy, straw procurement and transportation costs, comprehensive demand response, and carbon trading costs. It can calculate IESO profits and LA costs while meeting safety constraints.
[0050] The upper-level operational revenue maximization model is established using the following formula: In the formula, This represents IESO's profit; This is the total number of scenes; It is the probability of scenario p; This refers to the revenue from IESO selling energy to LA in scenario p. It is the penalty cost for wasted light and wind in scenario p; This refers to the demand response subsidy that IESO pays to LA in scenario p. It refers to the operation and maintenance costs of the equipment in scenario p; This refers to the cost of purchasing and transporting straw in scenario p. This represents the carbon trading cost in scenario p. , This indicates the electricity and heating prices set by IESO for LA. It is the operating and maintenance cost coefficient. . , The coefficient representing the penalty cost of curtailing solar and wind power is set at $82.88 / MW.
[0051] Specifically, the costs of purchasing and transporting straw are shown in the following formula: In the formula, It is the cost of transporting straw. This refers to the cost of purchasing straw.
[0052] The raw material for pyrolysis gasification devices is straw, typically sourced from nearby farmland. Assuming that straw within a 5-kilometer radius of the agricultural industrial park can be transported by farmers to the park, while straw beyond 5 kilometers can be transported by farmers to a straw collection center, which then transports it to the park, the straw transportation cost model is as follows: In the formula, The price is diesel fuel, set at US$1.13 per kilogram. The fuel consumption for a four-wheeled agricultural vehicle is calculated at $0.03 per kilometer. For truck fuel consumption, take $0.04 / km. This refers to the transportation distance from farmers to agricultural industrial parks. This refers to the transportation distance from farmers to straw collection centers. This refers to the transportation distance from the straw collection center to the agricultural industrial park. , The total number of farmers and farmers respectively. It is the distance from the farmer to the agricultural industrial park. This is the straw energy utilization rate, taken as 0.7. This refers to the amount of straw transported by farmers. It refers to the load-bearing capacity of a four-wheeled agricultural vehicle. It is the distance from farmer i to straw collection center k. It is the distance from the straw collection center (k) to the agricultural industrial park. This represents the amount of straw transported by the straw collection center k. It refers to the loading capacity of the truck. This indicates rounding.
[0053] The cost of purchasing straw is directly proportional to its purchase price and quantity, as shown in the following formula: In the formula, the straw purchase price It is $27.63 per ton.
[0054] Carbon trading costs are closely related to emission allowances, actual carbon emissions, and carbon trading prices. To encourage IESOs to reduce carbon emissions, a tiered carbon trading cost model was constructed, as shown below: In the formula, The carbon trading price is set at $13.82 per kilogram. Let 12 tons be the interval length. , This is the coefficient for punishment and reward, with values of 0.25 and 0.2 respectively. , It refers to the IESO emission allowance and actual carbon emissions.
[0055] In this embodiment, the off-grid IES carbon emission quota is shown in the following formula: CCS can capture carbon emissions from off-grid IES and provide carbon dioxide for MR to synthesize methane. Therefore, the actual carbon emissions can be expressed as follows: In the formula, The emission limit coefficient for heating is set at 0.392 kg / kW. The conversion coefficient from electrical energy to heat energy is taken as 1.67. , This is the carbon emission factor for combined heat and power (CHP), taken as -0.38 kg / kW and 0.0034 kg / kW. It is the carbon emission factor of PG, taken as 10 g / MJ.
[0056] In addition, the model must also meet the following constraints, including the sales prices of electricity and heat energy set by IESO, which must meet the following constraints: In the formula, , This represents the highest and lowest electricity prices. , These are the highest and lowest heating prices.
[0057] The power balance constraints that off-grid IES must meet are shown in the following formula: For step 104, based on the load model and energy sales price, a model for minimizing the energy purchase cost at the user end is established.
[0058] Based on the energy sales price set by the IESO, LA optimized energy demand with the goal of minimizing energy expenditure costs. Therefore, the lower-level energy purchase cost minimization model is shown in the following formula: In the formula, It is the cost of user comfort in scenario p; , These are the comfort cost coefficients for electrical energy and thermal energy, respectively, set at $1.1 / MW and $2.21 / MW.
[0059] For step 106, the upper-level operating revenue maximization model and the lower-level energy purchase cost minimization model are iteratively optimized based on the energy sales price constraint and the system balance constraint to calculate the optimal scheduling scheme of the agricultural park's integrated energy system.
[0060] Off-grid IES consists of two participants: IESOs (Industrial Engineering Stations) with energy supply, conversion, and storage capabilities, and LAs (Local Areas) with demand response capabilities. The game theory structure between the IESO and LA is as follows: Figure 3 As shown, IESO first sets the energy sales price for each time period of the day to maximize profits in Equation 20. Then, LA adjusts its energy demand based on the energy sales price to reduce energy expenditure costs in Equation 33. Furthermore, IESO will reset the energy sales price based on user-side demand response. The interaction between IESO and LA can be described as a leader-follower game. The game will repeat until neither staker can increase profits or reduce costs by unilaterally changing their operating strategies, reaching a game equilibrium.
[0061] like Figure 3As shown, IESO is the leader and LA is the follower. The two-level game model of the off-network IES is shown in the following formula. The above game model includes three elements: participants, strategies, and costs and profits. It is a set of participants, which can be represented as IESO and LA respectively. This represents the energy sales price decisions and output plans set by IESO for various devices. It is LA's demand response decision.
[0062] The above analysis shows that off-grid IES (Independent Ecosystem Oriented Components) involves multiple decision variables. The optimization process iterates over time, and the model solution method needs to consider both computational accuracy and efficiency. Traditional optimization algorithms require finding the optimal solution to the objective function, while the Differential Evolution (DE) algorithm does not require solving derivatives and constraints, reducing computational difficulty. However, like other organic intelligent algorithms, DE is prone to getting trapped in local optima. Therefore, this embodiment uses the Adaptive Differential Evolution (SHADE) algorithm based on success history to solve the profit maximization optimization problem of IESO. The cost optimization problem of LA (Local Allocation) is solved by using YALMIP to call the CPLEX solver.
[0063] Compared to DE, SHADE improves the mutation and crossover operators. Both operators are determined by an adaptive mechanism based on success history. SHADE sets up two history sets with H terms for the mutation and crossover operators, denoted as follows: and As shown in Table 1, SHADE preserves the H record to guide the two operators mentioned above to achieve adaptation during the search process.
[0064] Table 1 Historical memory of control parameters M Their average values are stored in the history memory and can be updated according to the following formula: In the formula, for The weighted average. express The weighted Lehmer average.
[0065] Initialize index h = 1. Whenever a new pair is stored in the history memory... He Shi h = h + 1. If no individual in the current iteration can produce a better test vector than its parent, the historical memory will not be updated. Even if the descendants... , Even if a set of poor values is included, the parent parameter will not be affected. Therefore, SHADE can improve the robustness of the algorithm and speed up convergence.
[0066] Each generation of individuals should undergo mutation and crossover according to the following formula: In the formula, It is a mutation operator, follows a Cauchy distribution, and has a location parameter of... The scale parameter is 0.1. It is a crossover operator, follows a normal distribution, and has a mean of . The standard deviation is 0.1.
[0067] Based on this, the optimization solution process in this embodiment includes: S61, initializing the preset total size U of energy sales price, the maximum number of iterations Z, the dimension of decision variables M, mutation and crossover operators, and the iteration convergence error. The initial population is obtained; S62. Use the adaptive differential evolution algorithm (SHADE) to select the optimal price from the initial population as the current energy purchase price for the user. S63. Calculate the current cost and current energy demand of the user based on the current energy purchase price, and calculate the current revenue of the energy end based on the current cost and current energy demand.
[0068] Specifically, AMIP is used to call the CPLEX solver to calculate the energy requirements of the LA (Landing Utility) and the cost of responding to user-side demands. LA's energy needs are passed on to IESO. IESO develops new output plans for various equipment and records current profits. And energy sales prices.
[0069] S64. Update the current energy purchase price using an algorithm that is adaptively iterated based on historical memory, and calculate the updated current cost on the user side and the updated current revenue on the energy side based on the updated current energy purchase price; S65. Determine whether the revenue after the energy-side update is greater than the revenue before the update. If so, update the local optimum based on the updated current cost, current revenue, and current energy purchase price; otherwise, do not update the local optimum. S66. Determine whether the difference in revenue and cost before and after adjacent updates are both less than the iteration convergence error. If so, stop the iteration and output the local optimum of the last update as the global optimum. Otherwise, repeat steps S64-S66 until the maximum number of iterations is reached.
[0070] The feasibility of the above method is verified by an example below.
[0071] This embodiment is based on an off-grid IES demonstration project in an agricultural industrial park. The scheduling cycle is 24 hours, with a time interval of 1 hour. A weather station provides wind and solar energy data. Typical renewable energy power generation scenarios are identified through clustering methods, such as... Figure 4 (a) and Figure 4 As shown in (b), the probabilities of the five scenarios are 0.25, 0.2, 0.11, 0.15, and 0.29, respectively. Figure 4 As shown in the five scenarios above, the power curve shape of WT varies significantly, with peaks and troughs occurring at any time of day. In the five scenarios, the power curve shape of photovoltaics shows almost no difference, with peaks concentrated between 12:00 and 14:00. The clustering results conform to natural laws.
[0072] The lower and upper limits of the park's electricity price are set at US$41.4 / 179.42 / MWh, and the upper and lower limits are set at US$27.6 / 69 / MWh. The installed capacity and operation and maintenance cost coefficients are shown in Table 2. The overall size of the IESO energy sales price is set to 12, the maximum number of iterations is set to 90, and the iteration convergence error is set to 0.001.
[0073] Table 2 Equipment installed capacity and operation and maintenance cost coefficients To demonstrate the rationality of the optimal scheduling strategy proposed in this embodiment, five schemes were designed for comparison and explanation. The factors considered in each scheme are shown in Table 3. Table 4 shows a comparison of the calculation results under different schemes.
[0074] Table 3 Scheme Design Description Table 4 Comparison of results under different schemes Comparing the results of Scheme 1 and Scheme 2, considering the uncertainties of renewable energy can make off-grid IES operation more scientific and efficient. It can reduce the IESO's need for straw procurement, lowering straw procurement and transportation costs by 12.5%. Furthermore, Scheme 2 increased IESO profits by 1.98% and reduced LA costs by 2.63%.
[0075] Comparing the results of Scheme 2 and Scheme 3, carbon trading costs could reduce IESO profits by 7.76% and increase LA costs by 1.53%. However, Scheme 3 reduces carbon emissions by 1,130 tons, reflecting the energy conservation and emission reduction concept in rural areas dominated by agriculture, and contributing to the green and low-carbon development of agricultural industrial parks.
[0076] By comparing the results of Schemes 3 and 4, reasonable mechanisms can be used to incentivize LA's demand responsiveness, such as demand responsiveness subsidies and user comfort costs. Taking demand responsiveness into account will reduce LA costs by 5.37%, increase IESO profits by 2.35%, and reduce carbon emissions by 1367 tons. Figure 5 As shown, considering the PMV index can ensure that the range of indoor temperature changes meets the requirements of human comfort, resulting in a smoother indoor temperature fluctuation. Therefore, considering the PMV index can effectively prevent the excessive conversion of other energy sources into heat energy, helping to balance user comfort and energy consumption costs.
[0077] Option 5 does not consider the game between IESO and LA, meaning IESO determines energy sales prices without taking LA's demand response into account. Leader-follower game theory helps explore LA's demand response potential, increasing carbon emission reductions and reducing carbon trading costs. Comparing the results of Option 4 and Option 5, although the cost of user comfort increased by $59.48, the demand response subsidy also increased by $132.2. Meanwhile, Option 4 reduced carbon emissions by 1824 tons, increased IESO profits by 5.02%, and reduced LA costs by 3.12%.
[0078] The iterative curves of IESO profit and LA cost are as follows: Figure 6 As shown, during the iteration process, both IESO's profit and LA's cost showed a gradual upward trend, reflecting the game between the two stakeholders. After 70 iterations, the game between IESO's profit and LA's cost reached equilibrium. Neither stakeholder can now further increase cost and profit by changing their respective independent strategies.
[0079] When the game reaches equilibrium, the energy sales price set by IESO and the energy demand of LA are as follows: Figure 7 As shown. During the periods of 11:00–12:00 and 17:00–22:00, peak electricity pricing helps incentivize LA to adjust energy demand and reduce energy consumption costs. Figure 8 As shown, demand response subsidies (LAs) can shift electricity load from peak hours to off-peak hours (e.g., 1:00 AM to 10:00 AM), reducing users' electricity costs. During dispatch, the total amount of load that can be shifted remains zero. Simultaneously, demand response subsidies can also guide LAs to reduce electricity load. The total electricity load after demand response is lower than the original electricity load, such as... Figure 7 As shown in (a). After obtaining the temperature change requirements of LA based on the PMV index, the required thermal energy for all time periods can be dynamically optimized. Figure 7 As shown in (b), although the total heat load demand has decreased slightly, the range of indoor temperature variation meets the requirements for physical comfort, such as Figure 5As shown. Therefore, the method proposed in this invention can align the profits of the IESO with the costs of the LA, and make the IESO's adjustment of energy sales prices consistent with the changing trends of LA energy demand.
[0080] The optimal dispatch results of electricity, heat and natural gas in off-grid IES are as follows: Figure 9 As shown. Off-grid IES prioritizes the use of renewable energy for power generation to reduce curtailment of wind and solar power. For example... Figure 9 As shown in (a), during the period from 00:00 to 07:00, the power load is supplied by WT and CHP, and the excess power can be provided to P2G for methane production. EES can maintain the power balance between power supply and consumption through charging and discharging. During the period from 08:00 to 09:00, renewable energy generation is insufficient to meet the power load demand, and part of the power load needs to be supplied by cogeneration and EES. During the period from 10:00 to 16:00, renewable energy generation can meet the power load demand, and the excess power can be used for methane production or stored in EES. During the peak power load period from 16:00 to 22:00, WT works with CHP and EES to meet the power load demand.
[0081] like Figure 9 As shown in (b), the heat load is supplied by CHP, HP, and HES. For the most part, CHP and HP can meet LA's heat load demand, and excess heat energy can be stored in HES. Between 18:00 and 22:00, when LA's heat load demand exceeds the supply capacity of CHP and HP, HES can release heat energy to maintain a heat energy balance between supply and demand.
[0082] like Figure 9 As shown in (c), most of the natural gas demand for cogeneration is met by the pyrolysis gasification unit. This allows for full utilization of straw resources in rural areas, reducing environmental pollution. The P2G operates according to the renewable energy generation curve. From 01:00 to 03:00, the P2G operates by consuming surplus electricity from renewable energy sources and, together with the pyrolysis gasification unit, meets the natural gas demand for cogeneration. From 11:00 to 16:00, renewable energy generation reaches its peak, and the P2G maintains high-power operation, meeting most of the natural gas demand for cogeneration. During dispatch periods, the GS can store excess natural gas and release it when renewable energy generation is insufficient to meet the power load demand of off-grid IES.
[0083] The applicability of SHADE is illustrated using Scheme 4 as an example. Table 5 compares the cost-profit and convergence results when using SHADE, DE, and Genetic Algorithm (GA) to solve game problems. Compared to DE and GA, SHADE achieves the best optimization results, with an IESO profit of $5611.07 and a LA cost of $5032.45. Furthermore, when using SHADE to solve the game model, it reduces the number of iterations by 202 and 235 compared to DE and GA, respectively. The time to solve the game model is also reduced by 142 and 89 seconds, respectively.
[0084] Table 5 Comparison of results for SHAED, DE, and GA The above data can verify the applicability of SHAED in solving the model proposed in this invention.
[0085] This invention provides a scheduling optimization device for an integrated energy system in an agricultural park, the device comprising: The first modeling module is used to establish equipment operation models on the energy side and load models on the user side based on the operation data of the integrated energy system of the agricultural park. The second modeling module is used to establish an upper-level operational revenue maximization model that includes straw procurement and transportation costs on the energy side, based on the system equipment operation model. The third modeling module is used to establish a model for minimizing the energy purchase cost at the user end based on the load model and energy sales price. The optimization module is used to iteratively optimize the upper-level operating revenue maximization model and the lower-level energy purchase cost minimization model based on energy sales price constraints and system balance constraints, and calculate the optimal scheduling scheme of the agricultural park's integrated energy system.
[0086] In this embodiment of the invention, establishing an equipment operation model for the energy sector based on the operational data of the integrated energy system in the agricultural park includes: establishing a probabilistic model for renewable energy power generation based on the probability distribution characteristics of wind speed and light intensity; establishing a linear gas production model for the pyrolysis gasification device based on the straw energy conversion efficiency; establishing a linear energy conversion model for heat pumps and cogeneration based on electrothermal conversion efficiency and cogeneration operation data; establishing an energy consumption model for the multi-stage chemical reaction of the electrochemical gasification device based on electrochemical principles and carbon capture technology; establishing a dynamic capacity model for the electric / thermal energy storage device based on the law of conservation of energy and charge / discharge efficiency characteristics; and establishing a seasonal gas storage model for the gas storage tank based on the operating parameters of the gas storage.
[0087] In this embodiment of the invention, a load model for the user end is established based on the operation data of the integrated energy system of the agricultural park, including: establishing a demand response model for flexible electrical load based on the load adjustability characteristics and the preset tiered incentive mechanism; and establishing a user experience-driven model for flexible thermal load based on the thermal comfort index and heat transfer data determined by the user end through predicted average voting.
[0088] In this embodiment of the invention, the upper-level operational revenue maximization model is established using the following formula: In the formula, This represents IESO's profit; This is the total number of scenes; It is the probability of scenario p; This refers to the revenue from IESO selling energy to LA in scenario p. It is the penalty cost for wasted light and wind in scenario p; This refers to the demand response subsidy that IESO pays to LA in scenario p. It refers to the operation and maintenance costs of the equipment in scenario p; This refers to the cost of purchasing and transporting straw in scenario p. This represents the carbon trading cost in scenario p.
[0089] In this embodiment of the invention, the lower-level energy purchase cost minimization model is established using the following formula: In the formula, It is the cost of user comfort in scenario p; , It is the comfort cost coefficient for electrical and thermal energy; This refers to the revenue from IESO selling energy to LA in scenario p. This refers to the demand response subsidy that IESO pays to LA in scenario p. The heat load represents time t; for The heat load at time t; This represents the electrical load after the demand response at time t; This represents the electrical load before the demand response at time t.
[0090] In this embodiment of the invention, the step of iteratively optimizing the upper-level operating revenue maximization model and the lower-level energy purchase cost minimization model based on energy sales price constraints and system balance constraints to calculate the optimal scheduling scheme for the agricultural park's integrated energy system includes: S61. Initialize the overall size of the preset energy sales price, as well as the maximum number of iterations, the dimension of decision variables, mutation and crossover operators, and the iteration convergence error to obtain the initial population; S62. Use the adaptive differential evolution algorithm to select the optimal price from the initial population as the current energy purchase price for the user. S63. Calculate the current cost and current energy demand of the user based on the current energy purchase price, and calculate the current revenue of the energy end based on the current cost and current energy demand. S64. Update the current energy purchase price using an algorithm that is adaptively iterated based on historical memory, and calculate the updated current cost on the user side and the updated current revenue on the energy side based on the updated current energy purchase price; S65. Determine whether the revenue after the energy-side update is greater than the revenue before the update. If so, update the local optimum based on the updated current cost, current revenue, and current energy purchase price; otherwise, do not update the local optimum. S66. Determine whether the difference in revenue and cost before and after adjacent updates are both less than the iteration convergence error. If so, stop the iteration and output the local optimum of the last update as the global optimum. Otherwise, repeat steps S64-S66 until the maximum number of iterations is reached.
[0091] It should be noted that the scheduling optimization device for the integrated energy system of agricultural parks provided in the above embodiments and the scheduling optimization method for the integrated energy system of agricultural parks belong to the same concept. The specific implementation process can be found in the method embodiments, which will not be repeated here.
[0092] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.
Claims
1. A scheduling optimization method for an integrated energy system in an agricultural park, characterized in that, The method includes: Based on the operational data of the integrated energy system in the agricultural park, establish an equipment operation model for the energy side and a load model for the user side; Based on the system equipment operation model, a model for maximizing upper-level operational revenue, including straw procurement and transportation costs, is established. Based on the load model and energy sales price, establish a model for minimizing the energy purchase cost at the user end. Based on the constraints of energy sales price and system balance, the upper-level operating revenue maximization model and the lower-level energy purchase cost minimization model are iteratively optimized to obtain the optimal scheduling scheme of the agricultural park's integrated energy system.
2. The method as described in claim 1, characterized in that, The process of establishing an equipment operation model for the energy sector based on operational data from the integrated energy system of the agricultural park includes: Based on the probability distribution characteristics of wind speed and solar intensity, a probabilistic model for renewable energy power generation is established. Based on the energy conversion efficiency of straw, a linear gas production model for the pyrolysis gasification device is established. Based on the electrothermal conversion efficiency and cogeneration operation data, a linear energy conversion model for heat pumps and cogeneration is established. Based on electrochemical principles and carbon capture technology, an energy consumption model for multi-stage chemical reactions in an electrochemical gas generator is established. Based on the law of conservation of energy and the characteristics of charge and discharge efficiency, a dynamic capacity model of an electric / thermal energy storage device is established. Based on the operating parameters of gas storage, a seasonal gas storage model for the gas storage tank is established.
3. The method as described in claim 1, characterized in that, Based on the operational data of the integrated energy system in the agricultural park, a load model for the user end is established, including: Based on the load's adjustable characteristics and the pre-set tiered excitation mechanism, a demand response model for flexible electrical loads is established. Based on the thermal comfort index and heat transfer data determined by users through predicted average voting, a user experience-driven model for flexible heat load is established.
4. The method as described in claim 1, characterized in that, The upper-level operational revenue maximization model is established using the following formula: In the formula, This represents IESO's profit; This is the total number of scenes; It is the probability of scenario p; This refers to the revenue from IESO selling energy to LA in scenario p. It is the penalty cost for wasted light and wind in scenario p; This refers to the demand response subsidy that IESO pays to LA in scenario p. It refers to the operation and maintenance costs of the equipment in scenario p; This refers to the cost of purchasing and transporting straw in scenario p. This represents the carbon trading cost in scenario p.
5. The method as described in claim 1, characterized in that, The lower-level energy purchase cost minimization model is established using the following formula: In the formula, It is the cost of user comfort in scenario p; , It is the comfort cost coefficient for electrical and thermal energy; This refers to the revenue from IESO selling energy to LA in scenario p. This refers to the demand response subsidy that IESO pays to LA in scenario p. The heat load represents time t; for The heat load at time t; This represents the electrical load following the demand response at time t. This represents the electrical load before the demand response at time t.
6. The method as described in claim 1, characterized in that, The iterative optimization of the upper-level operating revenue maximization model and the lower-level energy purchase cost minimization model based on energy sales price constraints and system balance constraints yields the optimal scheduling scheme for the agricultural park's integrated energy system, including: S61. Initialize the overall size of the preset energy sales price, as well as the maximum number of iterations, the dimension of decision variables, mutation and crossover operators, and the iteration convergence error to obtain the initial population; S62. Use the adaptive differential evolution algorithm to select the optimal price from the initial population as the current energy purchase price for the user. S63. Calculate the current cost and current energy demand of the user based on the current energy purchase price, and calculate the current revenue of the energy end based on the current cost and current energy demand. S64. Update the current energy purchase price using an algorithm that is adaptively iterated based on historical memory, and calculate the updated current cost on the user side and the updated current revenue on the energy side based on the updated current energy purchase price; S65. Determine whether the revenue after the energy-side update is greater than the revenue before the update. If so, update the local optimum based on the updated current cost, current revenue, and current energy purchase price; otherwise, do not update the local optimum. S66. Determine whether the difference in revenue and cost before and after adjacent updates are both less than the iteration convergence error. If so, stop the iteration and output the local optimum of the last update as the global optimum. Otherwise, repeat steps S64-S66 until the maximum number of iterations is reached.
7. A scheduling and optimization device for an integrated energy system in an agricultural park, characterized in that, The device includes: The first modeling module is used to establish equipment operation models on the energy side and load models on the user side based on the operation data of the integrated energy system of the agricultural park. The second modeling module is used to establish an upper-level operational revenue maximization model that includes straw procurement and transportation costs on the energy side, based on the system equipment operation model. The third modeling module is used to establish a model for minimizing the energy purchase cost at the user end based on the load model and energy sales price. The optimization module is used to iteratively optimize the upper-level operating revenue maximization model and the lower-level energy purchase cost minimization model based on energy sales price constraints and system balance constraints, and calculate the optimal scheduling scheme of the agricultural park's integrated energy system.
8. A computer device, characterized in that, The computer device includes a memory and a processor. The memory is used to store computer programs, and the processor is used to execute the computer programs stored in the memory to implement the steps of the method according to any one of claims 1-6.
9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the steps of the method described in any one of claims 1-6.
10. A computer program product, characterized in that, Includes a computer program, which, when executed by a processor, implements the steps of the method according to any one of claims 1-6.