Multi-energy flow collaborative optimization scheduling method and device and electronic equipment

Abstract

The application provides a multi-energy flow collaborative optimization scheduling method and device and electronic equipment, and relates to the field of micro-grid energy management and multi-energy flow collaborative optimization scheduling. In view of the problems that the separation scheduling of electricity, heat and carbon in the existing scheme leads to distortion of comprehensive cost, rough modeling of heat storage, rigid treatment of steam load, causes scheduling rigidity and insufficient energy supply stability, a multi-energy flow collaborative optimization scheduling method is provided. The method constructs a unified coupling model integrating power network, heat network and carbon emission flow, and combines real-time data to represent energy node balance and carbon emission characteristics. The method generates a long-time domain scheduling plan with the optimal comprehensive cost of fuel, power purchase, carbon emission and equipment loss as the target, and performs rolling correction based on model predictive control in a short time domain. When the power fluctuates, the electric heat pump, heat storage tank and heat flexible load are linked to respond. The scheme can improve the scheduling accuracy and flexibility, reduce the comprehensive operation cost and carbon emission, and improve the power supply and steam supply stability.

Description

Technical Field

[0001] This application relates to the field of microgrid energy management and multi-energy flow collaborative optimization scheduling, and in particular to a multi-energy flow collaborative optimization scheduling method, device and electronic equipment. Background Technology

[0002] Textile factories typically use a combination of boiler steam supply and microgrid power supply, and achieve coordinated operation of electricity and heat through electrothermal conversion, waste heat recovery and heat storage configuration.

[0003] However, existing dispatching schemes mostly revolve around the balance of electricity and heat, and carbon emissions and their costs are usually not uniformly incorporated into operational optimization. At the same time, the dynamic characteristics of thermal storage systems are described in a rather coarse manner, and steam loads are often treated as rigid demands, making it difficult to balance dispatching accuracy, economy, and operational stability in scenarios involving photovoltaic fluctuations and load changes.

[0004] Therefore, how to improve the accuracy of multi-energy flow coordinated scheduling in textile mill microgrids, while taking into account operating costs, carbon emissions, and energy supply stability, has become an urgent technical problem to be solved. Summary of the Invention

[0005] This application provides a multi-energy flow collaborative optimization scheduling method, apparatus, and electronic device to solve the aforementioned technical problems. This method addresses the operational scenario of a textile mill microgrid where electricity, heat, and carbon emissions are coupled. It focuses on the overall collaborative optimization of the multi-energy system, taking into account the economy, low-carbon emissions, and stability of the energy supply process. It comprehensively analyzes and dynamically schedules the operating status for future periods, thereby improving the accuracy of multi-energy flow collaborative scheduling and enhancing the overall operational performance under conditions of photovoltaic fluctuations and load changes.

[0006] Firstly, this application provides a multi-energy flow cooperative optimization scheduling method, the method comprising:

[0007] A multi-energy flow coupling model is established, which includes power networks, heat networks, and carbon emission flows. The multi-energy flow coupling model is used to dynamically describe the power balance relationship and carbon emission factor of each energy node through real-time operating data.

[0008] The mixed-integer programming problem is solved with the objective function of minimizing the overall operating cost to obtain the scheduling plan for the first time period in the future. The overall operating cost includes fuel cost, electricity purchase cost, carbon cost, and equipment depreciation cost.

[0009] During the execution of the scheduling plan, current actual operating data is collected, and the scheduling instructions in the future second time period in the scheduling plan are rolled and corrected through model prediction control using the current actual operating data as the initial state. The duration of the future second time period is less than the duration of the future first time period.

[0010] In the event of a detected power fluctuation, at least one of the following is adjusted in a coordinated manner: the power of the electric heat pump, the heat storage tank charging and discharging power, and the energy demand of the thermal elastic load, in order to respond to the power fluctuation and maintain a stable system energy supply.

[0011] The establishment of a multi-energy flow coupling model including power networks, thermal networks, and carbon emission flows includes:

[0012] The power dynamic balance constraint of the power network is established based on the photovoltaic output, the charging and discharging power of the energy storage unit, the diesel output, the grid switching power, and the power consumption of the electric heat pump, and the real-time state of charge of the energy storage unit is updated based on the dynamic equation of the energy storage state of charge.

[0013] The dynamic balance constraint of the thermal power of the thermal network is established based on the heating power of the electric heat pump, the waste heat recovery power, the heating power of the boiler, and the charging and discharging power of the thermal storage tank. The real-time total heat storage of the thermal storage tank is calculated based on the layered thermodynamic model of the thermal storage tank.

[0014] The instantaneous carbon emission intensity of the power load is determined by the weighted average of the carbon emission factors and their output proportions of photovoltaic power generation, energy storage unit charging and discharging, diesel power generation and grid power purchase.

[0015] The instantaneous carbon emission intensity of the steam load is determined by the weighted average of the carbon emission factors of the electric heat pump, waste heat recovery, and boiler, and their respective heat supply ratios; wherein, the carbon emission factor of the electric heat pump is determined by the ratio of the instantaneous carbon emission intensity to the energy efficiency coefficient of the electric heat pump.

[0016] The total system carbon emissions of the carbon emission stream are calculated based on the cumulative values ​​of the product of diesel power generation and diesel carbon emission factor, the product of grid exchange power and grid real-time carbon emission factor, and the product of boiler heating power and boiler carbon emission factor within the scheduling cycle.

[0017] In one possible embodiment, before solving the mixed-integer programming problem with the objective function of minimizing the overall operating cost, the method further includes:

[0018] Obtain the delay time, acceptable deviation of steam supply, and priority of each heat-consuming process;

[0019] The thermoelasticity coefficient of each heat-using process is determined based on the delay time, the acceptable deviation, and the priority, wherein the thermoelasticity coefficient is used to characterize the tolerance of the corresponding heat-using process to fluctuations in steam supply.

[0020] In one possible embodiment, the step of solving a mixed-integer programming problem with the objective function of minimizing the overall operating cost to obtain the scheduling plan for the first future time period includes:

[0021] Using the first future time period as the scheduling cycle, the scheduling cycle is discretized into multiple scheduling time periods according to a preset time step. The objective function is to minimize the sum of fuel cost, electricity purchase cost, carbon cost, energy storage unit life loss cost and thermal storage tank life loss cost. A day-ahead scheduling optimization model is then established.

[0022] The constraints of the day-ahead scheduling optimization model are determined, wherein the constraints include dynamic balance constraints of power output, dynamic balance constraints of thermal power output, operation constraints of energy storage units, operation constraints of diesel generators, operation constraints of thermal storage tanks, and thermoelastic constraints of steam load.

[0023] The day-ahead scheduling optimization model is solved using a mixed integer programming solver to obtain the scheduling plan for each device in the first future time period.

[0024] In one possible embodiment, the step of using the current actual operating data as the initial state and rollingly correcting the scheduling instructions in the future second time period of the scheduling plan through model predictive control includes:

[0025] Using the second future time period as the prediction time domain, the prediction time domain is divided into multiple control intervals according to a preset control cycle;

[0026] Using the current actual operating data as the initial state, and with the optimization objective of minimizing the sum of fuel cost, electricity purchase cost, power dispatch command change penalty term and thermal dispatch command change penalty term, an intraday rolling optimization model is established. The power dispatch command change penalty term is determined based on the power dispatch command deviation between the current time and the previous control interval, and the thermal dispatch command change penalty term is determined based on the thermal dispatch command deviation between the current time and the previous control interval.

[0027] Based on the constraints of the day-ahead scheduling optimization model, the constraints of the intraday rolling optimization model are determined.

[0028] At the arrival of each control cycle, a mixed integer programming solver is used to solve the intraday rolling optimization model, and the scheduling instructions obtained in the current control interval are executed as the corrected scheduling instructions. The rolling solution process is repeated at the arrival of the next control cycle.

[0029] In one possible embodiment, the step of adjusting at least one of the following in response to detected power fluctuations: the power of the electric heat pump, the heat storage tank charging / discharging power, and the energy demand of the thermally elastic load:

[0030] When a power shortage is detected, the power consumption of the electric heat pump is reduced and the heat release power of the heat storage tank is increased to compensate for the steam supply gap caused by the reduced power of the electric heat pump.

[0031] When a steam power deficit is detected, the heat release power of the thermal storage tank is increased first in response to the steam power deficit;

[0032] If there is still a steam supply gap after the heat storage tank has reached its maximum allowable heat release power, the steam supply of the corresponding heat-using processes will be reduced in order of decreasing thermoelasticity coefficients to prioritize the steam supply of heat-using processes with lower thermoelasticity coefficients.

[0033] In one possible embodiment, the method further includes:

[0034] Obtain the current photovoltaic output and power load;

[0035] When the photovoltaic output is greater than the sum of the power load and the minimum power consumption of the electric heat pump, the power consumption of the electric heat pump is increased to convert the surplus electrical energy into heat energy and store it in the heat storage tank.

[0036] When the sum of the photovoltaic output, the maximum energy storage unit charging and discharging power, and the maximum diesel output is less than the sum of the power load and the minimum power consumption of the electric heat pump, the power consumption of the electric heat pump is reduced and the heat release power of the heat storage tank is increased to meet the steam load. When the heat supply of the heat storage tank is insufficient, the boiler is started to supplement the steam supply.

[0037] In one possible embodiment, the method further includes:

[0038] Obtain the current grid electricity purchase price and determine whether the grid electricity purchase price exceeds a preset electricity price threshold;

[0039] When the preset electricity price threshold is exceeded, the energy storage unit is controlled to discharge and the diesel generator is started to generate electricity to replace the purchase of electricity from the grid;

[0040] Reduce the power consumption of the electric heat pump and meet the steam load demand based on the heat stored in the heat storage tank.

[0041] Secondly, this application provides a multi-energy flow cooperative optimization scheduling device, comprising:

[0042] The module is used to establish a multi-energy flow coupling model that includes power networks, thermal networks, and carbon emission flows. The multi-energy flow coupling model is used to dynamically describe the power balance relationship and carbon emission factor of each energy node through real-time operating data.

[0043] The scheduling module is used to solve a mixed integer programming problem with the objective function of minimizing the overall operating cost, and to obtain the scheduling plan for the first time period in the future. The overall operating cost includes fuel cost, electricity purchase cost, carbon cost, and equipment depreciation cost.

[0044] The scheduling module is also used to collect current actual operating data during the execution of the scheduling plan, and use the current actual operating data as the initial state to continuously correct the scheduling instructions in the future second time period in the scheduling plan through model prediction control, wherein the duration of the future second time period is less than the duration of the future first time period;

[0045] The control module is used to adjust at least one of the following when a power fluctuation is detected: the power of the electric heat pump, the heat storage tank charging and discharging power, and the energy demand of the thermal elastic load, in order to respond to the power fluctuation and maintain the stability of the system's energy supply.

[0046] The establishment of a multi-energy flow coupling model including power networks, thermal networks, and carbon emission flows includes:

[0047] The power dynamic balance constraint of the power network is established based on the photovoltaic output, the charging and discharging power of the energy storage unit, the diesel output, the grid switching power, and the power consumption of the electric heat pump, and the real-time state of charge of the energy storage unit is updated based on the dynamic equation of the energy storage state of charge.

[0048] The dynamic balance constraint of the thermal power of the thermal network is established based on the heating power of the electric heat pump, the waste heat recovery power, the heating power of the boiler, and the charging and discharging power of the thermal storage tank. The real-time total heat storage of the thermal storage tank is calculated based on the layered thermodynamic model of the thermal storage tank.

[0049] The instantaneous carbon emission intensity of the power load is determined by the weighted average of the carbon emission factors and their output proportions of photovoltaic power generation, energy storage unit charging and discharging, diesel power generation and grid power purchase.

[0050] The instantaneous carbon emission intensity of the steam load is determined by the weighted average of the carbon emission factors of the electric heat pump, waste heat recovery, and boiler, and their respective heat supply ratios; wherein, the carbon emission factor of the electric heat pump is determined by the ratio of the instantaneous carbon emission intensity to the energy efficiency coefficient of the electric heat pump.

[0051] The total system carbon emissions of the carbon emission stream are calculated based on the cumulative values ​​of the product of diesel power generation and diesel carbon emission factor, the product of grid exchange power and grid real-time carbon emission factor, and the product of boiler heating power and boiler carbon emission factor within the scheduling cycle.

[0052] Thirdly, this application provides an electronic device, including: a processor, and a memory communicatively connected to the processor;

[0053] The memory stores computer-executed instructions;

[0054] The processor executes computer execution instructions stored in the memory to implement the multi-energy flow cooperative optimization scheduling method as described in the first aspect.

[0055] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a computer, are used to implement the multi-energy flow cooperative optimization scheduling method as described in the first aspect.

[0056] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, can implement the steps of the multi-energy flow cooperative optimization scheduling method as described in any of the first aspects.

[0057] This application provides a multi-energy flow collaborative optimization scheduling method, device, and electronic equipment. By establishing a multi-energy flow coupling model that includes power networks, thermal networks, and carbon emission flows, and combining real-time operating data to dynamically describe the power balance relationship and carbon emission factor of each energy node, it can more accurately reflect the electro-thermal coupling and carbon emission correlation characteristics in the textile mill microgrid, thereby improving the scheduling modeling accuracy. By solving a mixed integer programming problem with the goal of minimizing the overall operating cost, and by incorporating fuel costs, electricity purchase costs, carbon costs, and equipment depreciation costs into the optimization, it can achieve a synergistic trade-off between system operating costs and carbon emission costs, thereby improving the overall economic efficiency and low-carbon operation level. By using model predictive control based on current actual operating data to roll over scheduling instructions for the short term in the future during the execution of the scheduling plan, and by responding to power fluctuations by using at least one of the following methods: electric heat pump power regulation, thermal storage tank charging and discharging regulation, and thermal elastic load scheduling, it can enhance the system's adaptability to photovoltaic fluctuations and load changes, thereby improving power supply stability and scheduling execution accuracy. Attached Figure Description

[0058] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0059] Figure 1 A flowchart illustrating a multi-energy flow cooperative optimization scheduling method provided in this application embodiment. Figure 1 ;

[0060] Figure 2 A flowchart illustrating a multi-energy flow cooperative optimization scheduling method provided in this application embodiment. Figure 2 ;

[0061] Figure 3 A flowchart illustrating a multi-energy flow cooperative optimization scheduling method provided in this application embodiment. Figure 3 ;

[0062] Figure 4 A flowchart illustrating a multi-energy flow cooperative optimization scheduling method provided in this application embodiment. Figure 4;

[0063] Figure 5 A flowchart illustrating a multi-energy flow cooperative optimization scheduling method provided in this application embodiment. Figure 5 ;

[0064] Figure 6 A schematic diagram of the structure of a multi-energy flow collaborative optimization scheduling device provided in an embodiment of this application;

[0065] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0066] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concepts of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0068] This application relates to the field of industrial microgrid energy management and multi-energy flow coordinated scheduling technology, and is applied to the scenario of combined power and steam supply in textile factories. Textile factories are typically equipped with photovoltaic, energy storage, diesel generators, grid access units, as well as electric heat pumps, thermal storage tanks, waste heat recovery devices, and boilers to meet the continuous power and stable steam supply needs of dyeing, setting, drying and other processes.

[0069] Existing energy supply schemes for textile factories mostly employ a coordinated operation of boiler steam supply and microgrid power supply. This involves using electrothermal conversion, waste heat recovery, and thermal storage devices to achieve partial coupling of electrical and thermal energy, and scheduling boiler start-up and shutdown, energy storage charging and discharging, and the operation of electric heating equipment according to predicted loads. The basic principle is to separately schedule power and thermal balance, allocating the output of various equipment within preset time periods.

[0070] The above-mentioned schemes have significant limitations in actual operation. First, electricity, heat, and carbon emissions are usually treated separately, and carbon emission intensity and carbon costs are not included in the unified dispatch process. When electricity prices, fuel prices, and carbon prices change, the dispatch results are difficult to reflect the true comprehensive cost. Second, thermal storage systems often use coarse models that do not fully describe temperature stratification, heat loss, and dynamics of heat charging and discharging, resulting in discrepancies between heat prediction and actual available heat. Third, steam loads are mostly treated as rigid demand, failing to differentiate the tolerance differences of different processes to steam supply fluctuations. When photovoltaic output fluctuates or heat load changes abruptly, problems such as dispatch rigidity, frequent activation of backup heat sources, and decreased energy supply stability can easily occur.

[0071] Based on the aforementioned technical challenges, a multi-energy flow coupling model incorporating power networks, heating networks, and carbon emission flows is adopted. This model integrates the power supply equipment, heating equipment, and carbon emission information in the textile mill within the same scheduling framework, enabling the synchronous correlation of power balance, heat allocation, and carbon cost accounting. The power network, heating network, and carbon management module of the textile mill's photovoltaic-storage-diesel microgrid energy management system thus form a unified data foundation.

[0072] Based on this, taking comprehensive operating cost as the scheduling target, we first obtain the plan for a longer period of time in the future, and then make rolling corrections to the instructions for a shorter period of time in combination with the current actual operating data. When power fluctuations are detected, we call on the power regulation of electric heat pumps, the heat storage tank charging and discharging regulation and the thermal elastic load scheduling to respond. This can improve the accuracy of multi-energy flow coordinated scheduling and take into account operating costs, carbon emissions and energy supply stability.

[0073] Figure 1 A flowchart illustrating a multi-energy flow cooperative optimization scheduling method provided in this application embodiment. Figure 1 .like Figure 1 As shown, the method includes:

[0074] S101: Establish a multi-energy flow coupling model that includes power networks, thermal networks, and carbon emission flows.

[0075] Among them, the multi-energy flow coupling model is used to dynamically describe the power balance relationship and carbon emission factor of each energy node through real-time operating data.

[0076] In this step, a multi-energy flow coupling model is used to characterize the coupling relationship between the power network, the heating network, and the carbon emission flow, and serves as the unified modeling basis for subsequent optimal scheduling. During integrated implementation, the power network, heating network, and carbon emission flow are configured within the same time-series framework, with data alignment and state updates performed at the same scheduling step size, enabling the joint expression of equipment operating states under different energy forms within the same model. Specifically, the executing entity can be a control platform, which communicates with field equipment. Real-time operating data includes at least the output information of each energy device, energy-consuming load information, and relevant parameters reflecting the system's operating state.

[0077] In one possible embodiment, the power network is described using node power balance relationships, establishing an electrical power conservation relationship at each scheduling moment; the thermal network is described using heat balance relationships, establishing a heat power conservation relationship at each scheduling moment. Carbon emission flows are used to track the carbon emission factors corresponding to the power and thermal networks, and to determine the corresponding carbon emissions based on the operating status of each energy node. To ensure the model's executability, carbon emissions are calculated discretely by time period, and aggregated at each moment to form a carbon emission state quantity that is updated synchronously with the power balance and heat balance. Based on the above analysis, by unifying the modeling of the power network, thermal network, and carbon emission flows, and introducing a real-time data-driven dynamic update mechanism, it is possible to synchronously express changes in power balance, heat balance, and carbon emissions within the same scheduling framework, providing a constraint basis for subsequent cost minimization solutions.

[0078] S102: Solve a mixed-integer programming problem with the objective function of minimizing the overall operating cost to obtain the scheduling plan for the first time period in the future. The overall operating cost includes fuel cost, electricity purchase cost, carbon cost, and equipment depreciation cost.

[0079] In this step, the overall operating cost is used as the cost set in the optimization objective to guide the scheduling plan in selecting specific equipment operation combinations and power allocation methods. Mixed-integer programming is used to simultaneously handle continuous and discrete decision variables, enabling the planning phase to complete global optimization within the forecasting time domain.

[0080] In practice, the first time period can be set according to actual dispatching needs, and the electricity price, fuel price, carbon price, and load forecast values ​​for each time period are mapped to the corresponding discrete time. In the objective function, fuel cost is determined by the relevant fuel consumption and the corresponding fuel unit price; electricity purchase cost is determined by the amount of electricity purchased from the grid and the electricity price; carbon cost is determined by the carbon emissions and carbon price for each time period; and equipment depreciation cost is determined based on changes in equipment operating status and usage intensity.

[0081] In one possible embodiment, the discrete decision variables in the mixed-integer programming model include equipment start-up / shutdown states and charging / discharging states; the continuous decision variables include the output power, purchased power, and load-side allocation of each energy device. Model constraints include at least dynamic balance constraints for electrical power, dynamic balance constraints for thermal power, carbon emission calculation constraints, and equipment operation constraints. State variables with mutually exclusive relationships can be expressed using binary variables.

[0082] In the specific solution process, the system first receives forecast data, future production plans, electricity price information, fuel price, and carbon price signals, and then forecasts the relevant output and load. The forecast results, along with the multi-energy flow coupling model established in the preceding steps, are then input into the solver. The solver outputs the scheduling plan for the first future time period according to the principle of minimizing the objective function. This scheduling plan includes at least the equipment power setpoints, purchased power, and relevant scheduling instructions for each time period. For continuous production scenarios, the planning results can also be used to generate an instruction table and sent to the field control system as the baseline trajectory for rolling control.

[0083] Based on the above analysis, it can be seen that by incorporating fuel costs, electricity purchase costs, carbon costs, and equipment depreciation costs into the same objective function and solving the mixed integer programming problem under multi-energy flow coupling constraints, the resulting future first-period scheduling plan can maintain a unified decision-making approach even when economic costs and carbon emission costs fluctuate simultaneously.

[0084] In one possible embodiment, considering the significant differences in the tolerance of various heat-consuming processes in a textile mill to fluctuations in steam supply—for example, the sizing process requires high steam pressure stability, while the drying process allows for a certain degree of steam volume fluctuation within a short period—this differentiated tolerance characteristic has a quantifiable and rankable physical basis. Therefore, to transform the steam load from a rigid demand into a graded and adjustable demand, and to make the scheduling results more consistent with the actual characteristics of continuous multi-process production in a textile mill, this embodiment introduces a steam load thermoelasticity quantification model. By obtaining the delay time, acceptable deviation of steam supply, and priority of each process, the thermoelasticity coefficient of each process is calculated to characterize its tolerance to fluctuations in steam supply. The specific steps for determining the thermoelasticity coefficient are as follows:

[0085] Obtain the delay time, acceptable deviation of steam supply, and priority of each heat-consuming process; determine the thermoelasticity coefficient of each heat-consuming process based on the delay time, acceptable deviation, and priority, whereby the thermoelasticity coefficient is used to characterize the tolerance of the corresponding heat-consuming process to fluctuations in steam supply.

[0086] Specifically, the energy management system reads the process parameters of each heat-consuming process from the production execution system, including at least: the allowable delay time (denoted as ) for the start of heating to be postponed, provided that the final product quality requirements are met. (Unit: minutes); the acceptable deviation in steam supply that does not affect product quality when the steam flow rate or pressure fluctuates briefly in this heat-using process (denoted as...). (unit: kilowatt); and the importance of the heat-using process in the production process, i.e., priority (denoted as ). In one embodiment, the priority uses an integer scale from 1 to 5, where 1 represents the highest priority (e.g., critical processes that cannot be interrupted) and 5 represents the lowest priority (e.g., auxiliary processes that can be moderately delayed). In another embodiment, a scale from 1 to 10 can also be used, with larger values ​​indicating higher priority; the mapping relationship just needs to be kept consistent in subsequent calculations. The priority can be obtained either manually preset and stored in the system database, or dynamically generated by the production planning module based on order delivery dates and equipment status.

[0087] Furthermore, the thermoelastic coefficient of each process is determined: Based on the above three parameters, the thermoelastic coefficient of each heat-using process is set as a dimensionless value. Its value is limited to the range [0,1]. A larger value indicates that the process has a higher tolerance for short-term steam supply fluctuations and can be preferentially delayed or moderately reduced in steam supply while meeting process boundaries; a smaller value indicates that the process needs to maintain a more stable steam supply. The formula for calculating this thermoelastic coefficient is:

[0088] .

[0089] in, Let be the delay time for process j. This is the maximum delay time among all heat-using processes, used to normalize the time dimension; The acceptable deviation in steam supply for process j. The steam demand for this process under rated operating conditions is given by dividing the two to obtain the relative proportion of the deviation. Let be the priority value of process j; the larger its reciprocal, the lower the priority, corresponding to higher flexibility; α, β, and γ are weighting coefficients, satisfying... This is used to balance the contribution of the three types of factors to elasticity. In one embodiment, the weighting coefficient is taken as... =0.35、 =0.35、 =0.30. In another embodiment, if a factory prioritizes production continuity, it can... Adjusted to 0.30 Adjusted to 0.40 The value was adjusted to 0.30 to emphasize the importance of steam deviation tolerance. The system calculates the thermoelastic coefficient for each process step according to the above formula and stores the results as input parameters for scheduling optimization.

[0090] After calculating the thermoelasticity coefficients of each process, the system incorporates these coefficients as adjustable weights on the steam load side into the objective function or constraints. Specifically, when the mixed-integer programming model decides on the steam supply allocation at each moment within the scheduling cycle, processes with higher thermoelasticity coefficients are allowed to have their supply reduced or heating delayed during periods of insufficient steam supply or peak electricity prices, while processes with lower thermoelasticity coefficients are given priority in ensuring their steam demand. The model thus transforms the steam load from a uniform rigid demand into a tiered elastic demand, and simultaneously considers the coordination between steam supply stability and load adjustability in the overall operating cost.

[0091] For example, a textile factory includes three heat-consuming processes: Process A is sizing, with a delay time of 0 minutes (non-delayable), an acceptable steam deviation of 2kW, and a priority of 1 (highest). Its calculated thermoelasticity coefficient is low (close to 0), indicating it is not adjustable. Process B is drying, with a delay time of 30 minutes, an acceptable steam deviation of 50kW, and a priority of 3. Its calculated thermoelasticity coefficient is high (e.g., 0.65), indicating it can be moderately delayed or reduced. Process C is setting, with parameters between the two. Based on the calculated thermoelasticity coefficient, the system prioritizes adjusting the steam supply to Process B in subsequent optimizations, thereby reducing overall operating costs without affecting critical processes.

[0092] By acquiring the delay time, acceptable steam supply deviation, and priority of each heat-consuming process, and using the aforementioned weighted normalization formula to calculate the thermoelasticity coefficient of each process, the time flexibility, steam supply deviation tolerance, and scheduling importance at the process level are uniformly mapped into quantifiable elasticity indicators. This method enables the mixed integer programming model to transform steam load from rigid demand to graded adjustable demand, thereby forming a scheduling result that better reflects the continuous production characteristics of multiple processes in textile mills. It allows the calculation of comprehensive operating costs to simultaneously consider steam supply stability and load adjustability, ensuring reliable steam supply to critical processes while fully utilizing the adjustment potential of non-critical processes to reduce energy consumption and carbon emission costs.

[0093] S103: During the execution of the scheduling plan, the current actual operating data is collected, and the scheduling instructions in the future second time period in the scheduling plan are rolled and corrected through model prediction control using the current actual operating data as the initial state. The duration of the future second time period is less than the duration of the future first time period.

[0094] In this step, model predictive control is used to perform online corrections to the aforementioned longer-term scheduling plan during the execution phase, addressing output fluctuations, load deviations, equipment efficiency changes, and thermal storage status deviations. The second future time period represents the short-term prediction window used for rolling optimization; its duration is shorter than the first future time period, allowing the system to continuously update near-end control commands without abandoning the overall planning direction. In specific implementation, current actual operating data is collected according to a preset sampling period, and model predictive control calculations are initiated according to a preset control period. Current actual operating data includes at least the actual output, energy storage status, heating status of key system equipment, and relevant parameters reflecting the load-side status.

[0095] In one possible implementation, model predictive control replaces the corresponding state variables in the plan with the current actual operating data, using it as the new initial boundary. Then, in a second future time period, it re-predicts the relevant output and load, and re-solves the constrained optimization problem. During rolling corrections, the solution object updates the relevant equipment power and load scheduling commands around the near-end time domain. In the updated control sequence, the control quantity corresponding to the current control cycle is sent to the field for execution, while the remaining control quantities are retained as prediction references and recalculated with newly acquired data when the next control cycle arrives. This forms a closed-loop rolling process of "prediction—optimization—execution—re-prediction".

[0096] To ensure the engineering feasibility of rolling corrections, this application retains operational constraints adapted to the field equipment at the model predictive control layer. For equipment with limited power variations and state transitions, corresponding dynamic constraints are retained; for energy storage and heating-related equipment, corresponding state and operational boundaries are retained; for carbon emission flows, the rolling correction stage continues to inherit the carbon emission modeling method from S101, ensuring that carbon emission changes are synchronously incorporated into control decisions. To reduce the impact of prediction errors on control results, the system can also correct the deviation of relevant prediction values ​​for the second future time period based on the prediction error of the previous control cycle. Based on the above analysis, it can be seen that by using current actual operating data as the initial state and implementing model predictive control in a second time period shorter than the first, long-term planning and short-term corrections can be linked, allowing dispatch commands to be continuously corrected according to the field conditions, thereby maintaining consistency between the energy supply process and the model state.

[0097] S104: In the event of a detected power fluctuation, at least one of the following is adjusted in conjunction with the power of the electric heat pump, the heat storage tank charging and discharging power, and the energy demand of the thermal elastic load to respond to the power fluctuation and maintain stable system energy supply.

[0098] In this step, power fluctuations include power deficits and surpluses in the power grid, as well as heating deviations or steam header pressure deviations in the heating network. Electric heat pump power regulation is used to achieve electrothermal coupling regulation by changing the power consumption of the heat pump; thermal storage tank charging and discharging regulation is used to provide a rapid buffer on the thermal energy side; and thermally flexible load scheduling is used to redistribute steam supply based on the load side's tolerance for steam fluctuations. In practice, the scheduling system continuously monitors the system's operating status, and determines that a power fluctuation has occurred when relevant indicators exceed preset thresholds.

[0099] In one possible embodiment, when a power fluctuation is detected, the control system can respond by adjusting at least one of the following methods based on the current electrothermal balance: lowering or raising the power of the heat pump, adjusting the charging or discharging state of the thermal storage tank, and adjusting the steam supply on the load side. In some scenarios, it can also coordinate with the operating status of other heating or power supply equipment. During the implementation of thermal storage tank charging / discharging regulation, the control system determines the executable charging / discharging range based on the status of the thermal storage device and its heat exchange capacity. During the implementation of thermal elastic load scheduling, the system adjusts the steam supply based on the load side's tolerance for steam fluctuations. After the response is executed, the scheduling system continues to collect new field data, determines whether the power fluctuation has been eliminated, and feeds back the actual execution results to the rolling optimization process in S103, ensuring consistency between short-term response and rolling correction. Based on the above analysis, it can be seen that when power fluctuations occur, the individual or combined execution of heat pump power adjustment, thermal storage tank charging / discharging regulation, and thermal elastic load scheduling can reconstruct the electrothermal power distribution relationship under the current coupling constraints, maintaining a coordinated state between power supply and steam supply.

[0100] In a further possible embodiment, considering that photovoltaic power generation output can drop sharply within seconds due to meteorological factors such as cloud cover, and steam load can also experience instantaneous fluctuations due to process switching or equipment start-up and shutdown, such power fluctuations on the order of seconds to minutes cannot be handled in a timely manner by day-ahead scheduling or intraday rolling optimization. Electric heat pumps have the ability to continuously adjust within a range of 30% to 100% of rated power with a response time of less than 5 seconds; the heat storage tank's heat release valve can fully open within 2 seconds with a response time of less than 10 seconds; and thermal elastic load scheduling can reduce steam supply in stages according to the tolerance of each process. These methods have the advantages of fast response speed, good adjustment continuity, and no impact on the main production processes. Therefore, in order to quickly smooth out fluctuations when power or steam power shortages occur, and to avoid triggering frequent start-up and shutdown of diesel generators or sudden increases in boiler load, this embodiment introduces a linkage response mechanism for the real-time adjustment layer. Upon detecting power fluctuations, at least one of the following is linked adjustment: electric heat pump power, heat storage tank charging and releasing power, and thermal elastic load energy demand, to respond to power fluctuations and maintain stable system energy supply. The corresponding specific implementation method is as follows:

[0101] When a power shortage is detected, the power consumption of the electric heat pump is reduced, and the heat release power of the thermal storage tank is increased to compensate for the steam supply gap caused by the reduced power of the electric heat pump. When a steam power shortage is detected, the heat release power of the thermal storage tank is increased first to respond to the steam power shortage. If a steam supply gap still exists after the heat release power of the thermal storage tank reaches the maximum allowable value, the steam supply of the corresponding heat-using process is reduced in order of the thermal elasticity coefficient from high to low, so as to prioritize the steam supply of heat-using processes with lower thermal elasticity coefficients.

[0102] In practical implementation, the system can adjust the compressor speed through the electric heat pump frequency converter to change the power consumption of the electric heat pump, and control the flow rate of the heat storage medium through the inlet and outlet regulating valves of the heat storage tank and the circulating pump, thereby changing the heat release power. The heat storage tank can adopt a layered heat storage structure, with upper, middle and lower temperature detection points set in the tank to estimate the available heat release in real time. When a steam power shortage is detected, the controller first reads the available heat of the heat storage tank and the current heat exchange capacity. If it does not reach the maximum allowable value, the circulating flow rate on the heat release side and the opening of the heat exchanger are increased to increase steam output. If the heat release power has reached the maximum allowable value but the steam is still insufficient, the scheduling module sorts the steam branch valves of the corresponding process from high to low according to the thermoelasticity coefficient, and performs graded closing or flow limit on the valves to reduce the steam supply for each process. The models of the above-mentioned electric heat pump, heat storage tank and valve components can be selected according to the project capacity. The selection of frequency converter electric heat pump and layered heat storage tank facilitates continuous power adjustment. In practical applications, other models of these components can also be selected, and this application does not limit this.

[0103] After the thermal storage tank capacity is exhausted, the heat-consuming processes are systematically reduced according to their thermoelasticity coefficients, ensuring that heat supply regulation is consistent with the importance of each process. This allows for tiered handling of steam shortages and prioritizes ensuring the supply of steam to processes with low thermoelasticity, thereby maintaining the stability of the textile mill's continuous steam supply.

[0104] For example, a large textile printing and dyeing factory experienced a sudden drop in total steam supply of 0.8MW due to a malfunction of its electric heat pump at a certain moment. The system detected a drop in steam pressure at t=0s, triggering an alarm and determining a steam power shortage. The system first prioritized increasing the heat release power of the heat storage tank, fully opening the heat release valve within 2 seconds, and increasing the steam supply by 0.5MW at t=5s. After determining that the heat release power of the heat storage tank had reached the maximum allowable value, there was still a 0.3MW steam supply gap. The system then prioritized the heat-using processes according to their thermoelasticity coefficients from highest to lowest (drying θ=0.70 is the highest, setting θ=0.50 is the second highest, dyeing θ=0.55, and others θ=0.15~0.30). It prioritized reducing the steam supply of the drying process, which has the highest thermoelasticity coefficient, by 0.3MW, by closing its steam branch valves to achieve graded flow control, while maintaining normal supply to dyeing, setting, and other processes.

[0105] This application provides a multi-energy flow collaborative optimization scheduling method. By integrating the power network, heating network, and carbon emission flow into a single scheduling model, and serially executing long-term mixed integer programming, short-term model predictive control, and a rapid power fluctuation response mechanism, the scheduling system can continuously adjust equipment output, power purchase strategy, thermal storage behavior, and load regulation behavior based on real-time operating status. This allows for the completion of comprehensive operating cost accounting, carbon emission constraints, and stable energy supply control within the same closed-loop process. Furthermore, upon detecting power fluctuations, a combination of methods is employed: rapidly changing the power consumption of electric heat pumps through frequency conversion regulation; quickly opening the heat release valve of the thermal storage tank to compensate for steam shortages; and progressively reducing steam supply to heat-consuming processes according to thermal elasticity coefficients from high to low. This achieves rapid responses to power and steam shortages at the second to minute level. This approach fully utilizes the rapid and continuous adjustment capability of electric heat pumps, the short-term energy storage buffer capability of thermal storage tanks, and the graded tolerance characteristics of thermal elastic loads. It smooths out power fluctuations without requiring the start-up of diesel generators or boilers, ensuring stable steam supply to critical processes with low thermal elasticity while effectively reducing the additional costs associated with frequent equipment start-ups and shutdowns. It should be understood that the above examples are for illustrative purposes only and are not intended to be limiting.

[0106] Figure 2 A flowchart illustrating a multi-energy flow cooperative optimization scheduling method provided in this application embodiment. Figure 2 This embodiment provides a detailed explanation of the specific steps involved in establishing a multi-energy flow coupling model. For example... Figure 2 As shown, the method includes:

[0107] S201. Establish dynamic power balance constraints for the power network based on photovoltaic output, energy storage unit charging and discharging power, diesel output, grid switching power, and electric heat pump power consumption, and update the real-time state of charge of the energy storage unit based on the dynamic equation of energy storage state of charge.

[0108] In this step, considering that the power network and thermal network are modeled uniformly by the industrial microgrid energy management system, possessing the advantages of data homogeneity and time-scale alignment, enabling collaborative monitoring and coupled analysis of multiple energy flows (electricity and heat), this embodiment introduces dynamic balance constraints for power output and dynamic equations for the state of charge of energy storage, as well as dynamic balance constraints for thermal output and a layered thermodynamic model of the thermal storage tank, respectively, to provide an accurate mathematical description of the power and thermal networks. Simultaneously, considering that photovoltaic power generation is a zero-carbon power source, diesel generators and boilers are carbon emission sources, and the carbon emission factor of grid-purchased electricity changes in real time, this embodiment introduces a weighted average formula for calculating instantaneous carbon emission intensity to accurately characterize the instantaneous carbon emission characteristics of the system under different combinations of power and heat sources, and calculates the total carbon emissions of the system by cumulative integration of each carbon emission source.

[0109] Specifically, a dynamic balance constraint for power output is first established. Taking sampling time t within a scheduling cycle as an example, the photovoltaic output, energy storage unit charging and discharging power, diesel output, grid switching power, and electric heat pump power consumption at that time together constitute the power input. The dynamic balance equation that the power output must satisfy is set as follows:

[0110] .

[0111] in, The photovoltaic output can be calculated in real time based on light intensity and temperature, or it can be directly acquired in real time through the power sensor of the photovoltaic array. The charging and discharging power of the energy storage unit, A positive value indicates charging, and a negative value indicates discharging. The amplitude is limited by , This represents the maximum charging and discharging power of the energy storage unit. The firewood exerts its power, and the firewood exerts its power. satisfy And constrained by the slope rate ; Power exchanged by the power grid A value greater than 0 indicates purchasing electricity, while a value less than 0 indicates feeding electricity back into the grid. For electrical load, The power consumption of the electric heat pump, and the power consumption of the electric heat pump Determined by system scheduling based on heat load demand; other power loads This includes non-adjustable loads such as lighting and air conditioning.

[0112] Secondly, the real-time state of charge (SOC) of the energy storage unit is updated. The energy storage unit is composed of a lithium battery pack, and its SOC... This indicates the percentage of current remaining battery power relative to the rated capacity. (At the sampling time) Based on the current charging and discharging power of the energy storage unit The next sampling time is updated using discretized equations. The discretized equation for the charged state is:

[0113] .

[0114] in, For charging and discharging efficiency, a value of 0.92 can be used; The sampling time interval; This refers to the rated capacity of the energy storage. Must meet Constraints are imposed to prevent overcharging and over-discharging.

[0115] Through the aforementioned power balance constraints and energy storage SOC dynamic equations, the system can match power supply and demand in real time and accurately grasp the remaining capacity status of energy storage units, providing a basis for subsequent scheduling decisions.

[0116] S202. Establish dynamic balance constraints of thermal power for the thermal network based on the heating power of the electric heat pump, the waste heat recovery power, the heating power of the boiler, and the charging and discharging power of the thermal storage tank, and calculate the real-time total heat storage capacity of the thermal storage tank based on the layered thermodynamic model of the thermal storage tank.

[0117] In this step, dynamic balance constraints for thermal power are established: the heat supply side includes the heating power of electric heat pumps, the heat recovery power of waste heat, the heating power of boilers, and the heat release power of thermal storage tanks; the heat demand side includes the steam load power, the heat charging power of thermal storage tanks, and the power loss of the heating network. The balance constraints that the thermal power must satisfy are as follows:

[0118] .

[0119] in, This refers to the heating power of the electric heat pump. For waste heat recovery power, For boiler heating capacity, The heat release power of the thermal storage tank. For steam load power, To charge the thermal storage tank with heat, This refers to power loss in the heating network. The heating power of the electric heat pump. ,in The coefficient of performance (COP) can range from 3.0 to 4.5, depending on the operating conditions of the electric heat pump. Waste heat recovery power... This includes recovered heat from photovoltaic modules, air conditioning systems, and air compressors. Boiler heating capacity. It serves as a backup heat source, activated when the heat pump's heating capacity is insufficient. (Heating network losses) The formula for calculation is: ,in This is the heat network loss coefficient. For water supply temperature, For ambient temperature, This represents the length of the pipe.

[0120] Secondly, the real-time total heat storage capacity of the thermal storage tank is calculated: the thermal storage tank adopts a vertical layered tank with an insulation layer. In one embodiment, the effective height of the tank is H, and it is uniformly divided into N layers along the height direction, where N=10; in another embodiment, the layers can also be divided according to the actual size of the tank and the temperature, with N=15 or 20 layers set according to the accuracy requirements. Each layer is regarded as a uniform temperature unit, and at sampling time t, the temperature of the i-th layer is... The stratified thermodynamic model of the thermal storage tank satisfies the following differential equation:

[0121] .

[0122] in To increase heat flow, For heat release flow, For the inlet temperature, For the outlet temperature, Assuming the density of water is 1000 kg / m³, The specific heat capacity is taken as 4.2 kJ / (kg·℃). The thermal conductivity is taken as 0.6 W / (m·K). It is a single-layer volume. The height is for a single layer. The thermal conductivity of the boundary layer is treated as an adiabatic boundary. The temperature of each layer is collected at the sampling time by temperature sensors arranged in layers, and the temperature of each layer is mapped to the enthalpy value of that layer, and then summed to obtain the total heat storage:

[0123] .

[0124] The summation range is i = 1 to N. During the heat charging and discharging process, the maximum heat charging power is... The maximum heat dissipation power is And satisfy the constraint that heat charging and discharging cannot occur simultaneously. .

[0125] Through the above-mentioned thermal power balance constraints and the layered thermodynamic model of the thermal storage tank, the system can match the thermal supply and demand in real time and accurately grasp the internal temperature distribution and total heat storage of the thermal storage tank, providing a basis for thermal scheduling.

[0126] S203. Determine the instantaneous carbon emission intensity of the power load based on the weighted average of the carbon emission factors and their output proportions of photovoltaic power generation, energy storage unit charging and discharging, diesel power generation and grid power purchase.

[0127] In this step, the instantaneous carbon emission intensity of the electricity load is determined: the carbon emission factor of photovoltaic power generation. The carbon emission factor of the energy storage unit during charging and discharging is 0 kg CO2 / kWh. The carbon emission factor of diesel power generation is 0 kg CO2 / kWh (the carbon emissions during charging are included in the charging power source). The carbon emission factor for electricity purchased from the grid is 0.78 kg CO2 / kWh (calculated based on diesel density of 0.84 kg / L, calorific value of 42.7 MJ / kg, and emission factor of 74.1 tCO2 / TJ). Obtained from the power grid dispatch center. The instantaneous carbon emission intensity of the power load is obtained by weighted averaging based on the instantaneous output ratio of each power source.

[0128] .

[0129] The meanings of the symbols are consistent with those in the aforementioned power balance equation, and will not be repeated here.

[0130] S204. Determine the instantaneous carbon emission intensity of steam load based on the weighted average of the carbon emission factors of electric heat pumps, waste heat recovery and boilers and their respective heat supply ratios.

[0131] The carbon emission factor of an electric heat pump is determined based on the ratio of instantaneous carbon emission intensity to the energy efficiency coefficient of the electric heat pump.

[0132] In this step, after determining the instantaneous carbon emission intensity of the electrical load, the instantaneous carbon emission intensity of the steam load is determined: the carbon emission factor of the electric heat pump. That is, the carbon emission factor of an electric heat pump is the instantaneous carbon emission intensity of the corresponding electrical load divided by the energy efficiency coefficient; the carbon emission factor of waste heat recovery... The carbon emission factor of the boiler is 0 kg CO2 / kWh. The instantaneous carbon emission intensity of the steam load is 0.26 kg CO2 / kWh (calculated based on a natural gas emission factor of 56.1 tCO2 / TJ and a boiler efficiency of 90%). A weighted average is then applied based on the instantaneous heat supply percentage of each heat source to obtain the instantaneous carbon emission intensity of the steam load.

[0133] .

[0134] The meanings of the symbols in the above formulas are consistent with the definitions mentioned above, and will not be repeated here.

[0135] S205. Calculate the total system carbon emissions of the carbon emission flow based on the cumulative values ​​of the product of diesel power generation and diesel carbon emission factor, the product of grid exchange power and grid real-time carbon emission factor, and the product of boiler heating power and boiler carbon emission factor within the scheduling cycle.

[0136] In this step, after calculating the instantaneous carbon emission intensity of the power load, the total system carbon emissions of the carbon emission stream are calculated using the following formula:

[0137] .

[0138] The integration interval is the entire scheduling cycle. The first three terms in the integral formula correspond to the instantaneous carbon emission rates of the three carbon emission sources: diesel generator, power grid, and boiler, respectively. In one embodiment, if the scheduling cycle is 24 hours, the carbon emission rates at each time point are summed with an hour as the time step. In another embodiment, the integration can also be discretized with a minute as the step step. The shorter the step step, the higher the calculation accuracy.

[0139] By calculating the carbon emission intensity of electricity, the carbon emission intensity of heat, and the total carbon emissions of the system, the system can output the carbon emission results on the electricity side, the heat side, and the total amount, providing a complete carbon emission constraint basis for subsequent scheduling optimization.

[0140] Simultaneously, by embedding power balance, thermal balance, and carbon emission accounting into the same model, the energy storage status, thermal storage status, and carbon emission status can be updated synchronously, thereby ensuring that the scheduling calculation results are consistent with the actual operating status of the textile factory's combined electric and steam power supply. Because a layered thermodynamic model is used on the thermal side, the estimate of the available heat in the thermal storage tank more closely reflects the actual temperature distribution within the tank. Furthermore, the total carbon emissions of the system are obtained by accumulating each item according to equipment output, which can be directly used as the basis for carbon accounting in integrated scheduling. This approach also enables the electric heat pump to establish a unified mapping between heating and carbon accounting, thus giving the multi-energy flow coupled model computable and traceable carbon management capabilities.

[0141] This application provides a multi-energy flow collaborative optimization scheduling method. By establishing dynamic balance constraints for power output and dynamic equations for the state of charge of energy storage, dynamic balance constraints for thermal power and a layered thermodynamic model of the thermal storage tank, and a weighted average method for calculating instantaneous carbon emission intensity and cumulative total carbon emissions, a unified coupled modeling of the three energy flows of electricity, heat, and carbon is achieved. This provides an accurate model foundation for subsequent long-term time-domain scheduling plan generation with the goal of optimal overall cost and short-term rolling correction based on model predictive control. Especially during power fluctuations, this model can adjust the power consumption of the electric heat pump, the heat charging and discharging power of the thermal storage tank, and the thermoelastic load in a coordinated manner, ensuring that the system achieves multi-energy collaborative optimization while meeting supply and demand balance and carbon emission constraints.

[0142] Figure 3 A flowchart illustrating a multi-energy flow cooperative optimization scheduling method provided in this application embodiment. Figure 3 This embodiment provides a detailed explanation of the specific steps for obtaining the scheduling plan for the first future time period. For example... Figure 3 As shown, the method includes:

[0143] S301. Taking the first future time period as the scheduling cycle, the scheduling cycle is discretized into multiple scheduling time periods according to a preset time step. The objective function is to minimize the sum of fuel cost, electricity purchase cost, carbon cost, energy storage unit life loss cost and thermal storage tank life loss cost. A day-ahead scheduling optimization model is then established.

[0144] In this step, considering that the mixed-integer programming solver has the ability to handle multiple types of decision variables (including binary start-stop variables and continuous power variables) and complex constraints, it can search for the global optimal solution given the forecast data. Simultaneously, considering the irreversible lifespan decay of energy storage units and thermal storage tanks during frequent charging and discharging, incorporating lifespan loss costs into the objective function can more realistically reflect the impact of long-term equipment operation on overall economic efficiency. Furthermore, the introduction of the thermoelastic constraint on steam load allows the model to fully utilize the adjustment potential of each process while satisfying the process boundary. Therefore, in order to obtain the optimal scheduling scheme for the first future time period under multiple constraints of electricity and heat supply and demand balance and equipment operation safety, this embodiment uses the first future time period as the scheduling cycle, discretizes it into multiple scheduling time periods according to a preset time step, establishes a day-ahead scheduling optimization model, and uses a mixed-integer programming solver to solve the model.

[0145] Specifically, the scheduling cycle is defined as the first future time period. The scheduling cycle is discretized into multiple scheduling periods according to a preset time step. The first future time period and the preset time step can be selected based on the actual application scenario. For example, if the scheduling cycle is set to 24 hours and the preset time step is set to 1 hour, the scheduling cycle is discretized into 24 scheduling periods, each corresponding to an integer index t, t=1,2,…,24. If the system requires high prediction accuracy or experiences severe load fluctuations, the time step can be set to 30 minutes, discretizing the scheduling cycle into 48 scheduling periods. Shorter steps result in higher scheduling accuracy, but also increase the computational burden. The system selects an appropriate time step based on actual operational needs and control granularity.

[0146] Furthermore, the objective function is to minimize the sum of fuel cost, electricity purchase cost, carbon cost, energy storage unit lifespan depreciation cost, and thermal storage tank lifespan depreciation cost. The expression for this objective function is as follows:

[0147] .

[0148] in, The fuel cost is calculated as follows: ,in The diesel generator is used for power generation, and a, b, and c are fuel cost coefficients, which are obtained by fitting the diesel generator's fuel consumption characteristic curve.

[0149] The formula for calculating the cost of electricity is: ,in, For power exchange with the power grid, For time-of-use electricity pricing, For time interval step, A value greater than 0 indicates a positive electricity purchase cost. A value less than 0 indicates electricity sales, in which case this item represents negative revenue.

[0150] The carbon cost is calculated using the following formula: .in, Let be the total carbon emissions of the system during time period t. For carbon price;

[0151] The formula for calculating the lifespan depreciation cost of an energy storage unit is as follows: .in, The charging and discharging power of the energy storage unit, This is the energy storage unit lifetime degradation factor, which characterizes the accelerated degradation effect of high-power charging and discharging on battery life.

[0152] The formula for calculating the depreciation cost of a thermal storage tank over its lifespan is as follows: .in, and These are the thermal storage tank's charging power and thermal storage tank's dissipation power, respectively. This is the lifespan reduction factor for the thermal storage tank. The greater the heat charging and discharging power of the thermal storage tank, the greater the thermal stress impact on the tank, and the more significant the lifespan reduction.

[0153] S302. Determine the constraints of the day-ahead scheduling optimization model.

[0154] The constraints include dynamic balance constraints for electrical power, dynamic balance constraints for thermal power, operational constraints for energy storage units, operational constraints for diesel generators, operational constraints for thermal storage tanks, and thermoelastic constraints for steam load.

[0155] Energy storage unit operating constraints include at least: energy storage unit SOC constraints, i.e. To prevent overcharging and over-discharging; energy storage unit power constraint, i.e. ;

[0156] Diesel generator operation constraints, namely , This represents the maximum gradeability of the diesel engine. The time step size is used to ensure that the output variation between adjacent time periods does not exceed physical limits.

[0157] Operating constraints for thermal storage tanks include at least: upper and lower limits on the amount of heat stored, i.e. The thermal storage tank's charging power constraint, i.e. The heat release power constraint of the thermal storage tank, i.e. ;

[0158] The steam load thermoelastic constraint is used to define the response boundary of deferred processes to fluctuations in steam supply within permissible deviations. The specific expression is the actual supply... ,in Let j be the thermoelastic coefficient of the heat-using process. For the steam demand of this process in time period t, this constraint ensures that the actual steam supply of each process in any time period is not lower than its lower limit of demand, that is, it allows for a moderate reduction in supply within the range corresponding to the thermoelasticity coefficient, but cannot exceed the minimum steam supply allowed by the process.

[0159] The dynamic balance constraints of electrical power and thermal power have been described in detail in the foregoing embodiments and will not be repeated here.

[0160] S303. A mixed integer programming solver is used to solve the day-ahead scheduling optimization model to obtain the scheduling plan for each device in the first time period in the future.

[0161] In this step, the system first obtains the time-of-use electricity price for each time period. Gas price (already included in fuel cost coefficients a, b, and c), carbon price The predicted values ​​of power load, photovoltaic output, and steam load, as well as the initial states of each device (including the initial SOC of energy storage, the initial temperatures of each layer of the thermal storage tank, and the initial start-up and shutdown states of the diesel generator, etc.), are input into the day-ahead scheduling optimization model. Then, for the nonlinear terms in the layered thermodynamic model of the thermal storage tank, a piecewise linearization method is used in one embodiment to convert the nonlinear constraint into a multi-segment linear constraint, thereby transforming the original mixed-integer nonlinear programming problem into a mixed-integer linear programming problem, which is then solved using a solver. The solver searches for the optimal combination of binary start-up and shutdown variables and continuous power variables that minimizes the objective function within the feasible region that satisfies all constraints. Finally, it outputs the scheduling plan for each device in the first time period of the future. This scheduling plan includes the photovoltaic power consumption, energy storage unit charging and discharging power and state of charge trajectory, diesel generator start-up and shutdown status and output curve, grid power purchase and sale, electric heat pump power consumption, boiler heating power, thermal storage tank charging and discharging power and heat storage change trajectory, and steam supply allocation scheme for each process.

[0162] By incorporating fuel costs, electricity purchase costs, carbon costs, and the lifespan depreciation costs of energy storage units and thermal storage tanks into the objective function, and combining this with the thermoelastic constraints of electricity, heat, energy storage units, diesel generators, thermal storage tanks, and steam loads for joint solution, the resulting scheduling plan can achieve coordinated allocation of multiple energy flows within the same optimization framework, ensuring that the equipment output in the first time period is consistent with load demand, carbon emission accounting, and equipment lifespan consumption.

[0163] This application provides a multi-energy flow collaborative optimization scheduling method. It establishes a day-ahead scheduling optimization model with the objective function of minimizing the sum of fuel cost, electricity purchase cost, carbon cost, energy storage unit lifespan depreciation cost, and thermal storage tank lifespan depreciation cost. This model includes constraints on dynamic balance of power output, dynamic balance of thermal power output, operation of energy storage units, diesel generator operation, thermal storage tank operation, and thermoelasticity of steam load. A mixed-integer programming solver is used to solve the model, achieving global optimization of comprehensive operating costs while satisfying the supply and demand balance of multiple energy flows and ensuring equipment operational safety. This approach enables the generated scheduling plan to fully coordinate multiple decisions such as photovoltaic absorption, energy storage charging and discharging, diesel generator start-up and shutdown, electricity purchase and sales, and heat supply. Furthermore, it quantifies the adjustability of steam load as a thermoelastic constraint and incorporates it into the optimization process. This significantly reduces the comprehensive operating costs of fuel, electricity purchase, and carbon emissions while ensuring the reliability of steam supply to key processes.

[0164] Figure 4 A flowchart illustrating a multi-energy flow cooperative optimization scheduling method provided in this application embodiment. Figure 4 This embodiment provides a detailed explanation of the specific steps for revising the scheduling instructions for the second future time period in the scheduling plan. For example... Figure 4 As shown, the method includes:

[0165] S401. Using the second future time period as the prediction time domain, the prediction time domain is divided into multiple control intervals according to the preset control cycle.

[0166] In this step, during the intraday correction phase, considering the advantage of Model Predictive Control (MPC) in iterative online optimization based on the current measured state, it can utilize real-time feedback information to solve the problem in a finite time domain, thereby effectively suppressing scheduling deviations caused by the accumulation of day-ahead prediction errors. Simultaneously, considering that drastic changes in scheduling commands between adjacent control intervals can lead to frequent equipment start-ups and shutdowns and sudden power fluctuations, increasing mechanical wear and thermal stress shocks, reducing equipment lifespan, and affecting system stability, this embodiment introduces an intraday rolling optimization model based on Model Predictive Control to adapt to real-time changes in photovoltaic output, steam load, and carbon prices while tracking the day-ahead scheduling plan and suppressing frequent fluctuations in scheduling commands for each device. Using the current actual operating data as the initial state, the optimization objective is to minimize the sum of fuel costs, electricity purchase costs, carbon emission costs, and penalties for changes in power and thermal scheduling commands. The model solves and executes the corrected scheduling commands at the end of each control cycle.

[0167] Specifically, a second future time period is determined, and this second future time period is used as the prediction time domain. The prediction time domain is then divided into multiple control intervals according to a preset control cycle. The second future time period and the prediction time domain can be determined based on the actual application scenario.

[0168] In one embodiment, the prediction time domain, i.e., the second future time period, is set to 4 hours, and the preset control period is set to 15 minutes. This divides the prediction time domain into 16 control intervals, each corresponding to a decision step size, with indices k=1,2,…,16. In another embodiment, if the system has higher real-time requirements or the equipment requires faster adjustment and response, the prediction time domain can be set to 2 hours, and the control period to 10 minutes, resulting in 12 control intervals. Shorter control periods allow for more frequent responses to disturbances, but the computational burden increases accordingly. The control time domain comprises the first few control intervals within the prediction time domain. In one embodiment, this is set to 1 hour (i.e., 4 control intervals), indicating that the model only issues and executes scheduling instructions within the current control period; scheduling instructions in subsequent time domains are only used for prediction optimization and not directly executed.

[0169] S402. Using the current actual operating data as the initial state, and with the optimization objective of minimizing the sum of fuel cost, electricity purchase cost, carbon emission cost, penalty term for changes in power dispatch instructions, and penalty term for changes in heat dispatch instructions, establish an intraday rolling optimization model.

[0170] Among them, the penalty term for changes in power dispatch instructions is determined based on the deviation of power dispatch instructions between the current time and the previous control interval, and the penalty term for changes in thermal dispatch instructions is determined based on the deviation of thermal dispatch instructions between the current time and the previous control interval.

[0171] Specifically, taking the current actual operating data as the initial state, and minimizing the sum of fuel cost, electricity purchase cost, carbon emission cost, penalty term for changes in power dispatch instructions, and penalty term for changes in heat dispatch instructions as the optimization objective, the expression of this objective function is:

[0172] .

[0173] Where N is the total number of control intervals in the prediction time domain, t is the current time, and t+k represents the k-th control interval after the current time. For fuel costs, For electricity purchase costs, For carbon costs, the definitions of these three items are consistent with those in the day-ahead scheduling optimization model; This refers to the change in power dispatch instructions, representing the deviation between the current moment and the power dispatch instructions (including diesel generator output, energy storage charging and discharging power, grid switching power, and electric heat pump power consumption) in the previous control interval. This is the sum of squares of the deviation. This represents the deviation of thermal dispatch commands, indicating the vector of deviation between the current time and the thermal dispatch commands (including boiler heating power, thermal storage tank charging and releasing power, etc.) of the previous control interval. This is the sum of squares of the deviation vector; and These are the weighting coefficients for the penalty terms for changes in power dispatch commands and changes in thermal dispatch commands, respectively. Both are positive numbers, used to adjust the degree of suppression for command jumps. In one embodiment, and The value is determined empirically. Take 0.01, The weight is set to 0.008; in another embodiment, the optimal weight can also be determined by system identification or online adaptive adjustment to balance the trade-off between tracking performance and smoothness.

[0174] S403. Based on the constraints of the day-ahead scheduling optimization model, determine the constraints of the intraday rolling optimization model.

[0175] In this step, various constraints from the day-ahead scheduling optimization model are inherited and mapped to the feasible region within each control interval. Specifically, these constraints include at least: dynamic balance constraints for power output, dynamic balance constraints for thermal power output, operational constraints for energy storage units, operational constraints for diesel generators, operational constraints for thermal storage tanks, and thermoelastic constraints for steam load. The form of these constraints is consistent with that in the day-ahead scheduling optimization model, but all variables are replaced with the variables corresponding to control interval k, and the initial states (such as the initial state of charge of energy storage, the initial temperature of each layer of thermal storage tanks, and the initial start-up and shutdown states of diesel generators) are assigned values ​​based on the real-time operating data actually collected in the current control cycle.

[0176] S404. At the arrival of each control cycle, a mixed integer programming solver is used to solve the intraday rolling optimization model, and the scheduling instructions obtained in the current control interval are executed as the corrected scheduling instructions. The rolling solution process is repeated at the arrival of the next control cycle.

[0177] In this step, at the arrival of each control cycle, the system first collects the actual operating data for the current moment, including the actual output of photovoltaic power, the actual state of charge of the energy storage unit, the actual temperature of each layer of the thermal storage tank, the actual operating status of the diesel generator, the measured value of the grid exchange power, and the measured values ​​of the power load and steam load. This data is used as the initial state input for the intraday rolling optimization model. Then, combining the predicted values ​​of photovoltaic output, steam load, and carbon price for each control interval in the prediction time domain, a complete optimization problem is constructed and solved using a mixed-integer programming solver. The solver searches for the optimal combination of binary start-stop variables and continuous power variables that minimizes the objective function within the feasible region that satisfies all constraints. It outputs the discrete start-stop status and continuous output value of each device in the current control interval, and writes the result as a corrected scheduling instruction into the execution layer for execution. After the current control interval ends, the system waits for the arrival of the next control cycle, at which time the latest actual operating data is collected again, the system rolls forward one control cycle, updates the initial state, and repeats the above solution process.

[0178] In this way, the intraday rolling optimization model can combine the day-ahead plan with the real-time status to continuously revise the scheduling instructions in the next second time period. This ensures that the revision results are always constrained by the day-ahead constraint boundary, while reducing the magnitude of instruction mutations on the power and heat sides. This improves the tracking consistency and execution stability of multi-energy flow coordinated scheduling, and makes the overall operating cost and carbon emission control results more consistent with actual operating conditions.

[0179] This application provides a multi-energy flow collaborative optimization scheduling method. Starting with current operational data as the initial state, it establishes an intraday rolling optimization model with the objective function of minimizing the sum of penalties for changes in power dispatch commands and thermal dispatch commands. A mixed-integer programming solver is used to solve the model in each control cycle, enabling online correction of dispatch commands based on model predictive control. This method allows the corrected dispatch commands to adapt promptly to real-time changes in photovoltaic output, steam load, and carbon prices. Simultaneously, the penalty terms effectively suppress frequent jumps in power and thermal dispatch commands between adjacent control intervals, reducing overall operating costs while ensuring stable system operation.

[0180] Figure 5 A flowchart illustrating a multi-energy flow cooperative optimization scheduling method provided in this application embodiment. Figure 5 .like Figure 5 As shown, the method includes:

[0181] S501, Obtain the current photovoltaic output and power load.

[0182] In this step, the energy management strategy considers that the electric heat pump has the characteristic of converting electrical energy into heat energy, and its power consumption can be continuously adjusted within a certain range. Furthermore, the thermal storage tank has a long heat retention capacity and a large heat capacity, enabling it to store heat energy for a long time and release it when needed. This has a natural advantage in realizing the transfer of electrical and thermal energy across time scales. Simultaneously, the boiler, as a backup heat source, can be quickly started to compensate for the steam gap when the thermal storage tank's heat supply is insufficient. Therefore, in order to maximize the utilization of surplus electricity and avoid curtailment when photovoltaic output is sufficient, and to prioritize ensuring the balance of power supply and demand and the stability of steam load when photovoltaic output is insufficient, this embodiment introduces an active power adjustment mechanism for the electric heat pump based on the real-time difference between photovoltaic output and power load. This mechanism converts surplus electricity into heat energy storage by increasing the power consumption of the electric heat pump, or reduces the power consumption of the electric heat pump and supplements it with heat release from the thermal storage tank and boiler supplementation when power is insufficient, thus achieving coordinated electricity-heat scheduling.

[0183] Specifically, the system obtains the photovoltaic output at the current moment through the sampled values ​​of the photovoltaic inverter. The current power load is obtained through power load metering values. Simultaneously, the minimum power consumption of the electric heat pump is read from the system database. and maximum power consumption of electric heat pump Maximum energy storage unit charging and discharging power , maximum diesel power Parameters such as the current available heat supply of the thermal storage tank and the boiler standby status are used as input data for subsequent judgment and decision-making. In one embodiment, the sampling period is set to 1 second to quickly track changes in photovoltaic power and load; in another embodiment, it can also be set to 5 seconds according to the dynamic response requirements of the system. The higher the sampling frequency, the faster the response, but the requirements for communication and computing resources are also correspondingly increased.

[0184] S502. When the photovoltaic output is greater than the sum of the power load and the minimum power consumption of the electric heat pump, the power consumption of the electric heat pump is increased to convert the surplus electrical energy into heat energy and store it in the heat storage tank.

[0185] In this step, the system determines whether the current condition is met: the photovoltaic output is greater than the sum of the electrical load and the minimum power consumption of the electric heat pump. Is this condition met? If it is, it means that the current photovoltaic output has a surplus after meeting the power load and the minimum power consumption of the electric heat pump. The system then increases the power consumption of the electric heat pump to convert the surplus electrical energy into heat energy.

[0186] The specific operation is as follows: The controller sends a command to the electric heat pump inverter to increase the operating frequency of the electric heat pump compressor, thereby adjusting the power consumption of the electric heat pump upward from the current value. The increase is determined based on the amount of surplus power, but does not exceed the maximum power consumption of the electric heat pump. The difference between the current power consumption and the energy consumption. The heat generated by the electric heat pump is sent to the heat storage tank through the heat exchange circuit, and stored in the heat storage tank in the form of hot water or steam, completing the transfer and storage of surplus electrical energy into heat energy.

[0187] For example, a large printing and dyeing textile factory has a photovoltaic output of 2.5MW and an electrical load (excluding the electric heat pump) of 0.5MW during the period from 12:00 to 13:00. The minimum power consumption of the electric heat pump is 0.5MW. The judgment condition is... Once established, the system determines that it is a period of surplus photovoltaic power. The power consumption of the electric heat pump is increased from the minimum of 0.5MW to the full power of 2.0MW, increasing the power consumption by 1.5MW. This power is converted into heat energy and stored in the heat storage tank through the heat exchange loop (the heat storage tank is charged with 2.0MW). At the same time, the energy storage unit is charged with 0.5MW. The remaining 0.5MW of surplus power is fed back to the grid for profit, achieving zero curtailment of solar power.

[0188] S503. When the sum of photovoltaic output, maximum energy storage unit charging and discharging power, and maximum diesel output is less than the sum of power load and minimum electric heat pump power consumption, reduce the electric heat pump power consumption and increase the heat release power of the heat storage tank to meet the steam load, and start the boiler to supplement steam supply when the heat supply of the heat storage tank is insufficient.

[0189] In this step, the system determines whether the sum of photovoltaic output, maximum energy storage unit charging / discharging power, and maximum diesel output is less than the sum of electrical load and minimum electric heat pump power consumption; that is, the judgment condition is met. Is this condition met? If it is, it means that even under extreme conditions where the energy storage unit operates at maximum discharge power and the diesel generator operates at maximum output, it is still impossible to meet the sum of the current power load and the minimum power consumption of the electric heat pump. The system then enters a power deficit response mode. In this mode, the controller first reduces the power consumption of the electric heat pump, adjusting it towards the direction of minimum power consumption, in order to release the capacity on the power side and reduce the total power load demand.

[0190] Simultaneously, the system increases the heat release power of the thermal storage tank by widening the outlet regulating valve and increasing the frequency of the heat release-side circulating pump, thereby increasing the heat output from the thermal storage tank to the heating network to maintain stable steam load. If the heat release capacity of the thermal storage tank is still lower than the steam load demand, the controller sends a boiler start-up command, and the boiler burns natural gas or oil to generate steam, which is then used to supplement the shortfall via the steam header. In one embodiment, the available heat capacity of the thermal storage tank is estimated in real time by temperature detection points at the upper, middle, and lower layers of the tank. If the available heat capacity is greater than the shortfall, the thermal storage tank will fully compensate for the shortfall; otherwise, the boiler will supplement the remaining shortfall. In another embodiment, if the boiler's response from a cold start is slow, the system can predict the duration of the power deficit in advance and pre-start the boiler to enter a hot standby state before the deficit occurs, thereby shortening the actual response time for supplementing steam.

[0191] This implementation method utilizes electric heat pumps as adjustable loads to absorb excess electricity and convert it into heat energy when there is a surplus of photovoltaic power. During periods of power shortage, it maintains steam supply by reducing the power consumption of the heat pumps and coordinating with the thermal storage tank to release heat and the boiler to supplement heat. This allows the power and heat sides to respond collaboratively to power fluctuations at any given moment. Consequently, surplus electricity can be utilized locally, and a continuous steam supply can be maintained even when thermal storage is insufficient. This approach is suitable for real-time scheduling in scenarios involving combined electricity and steam power supply in textile factories.

[0192] S504. Obtain the current grid purchase price of electricity and determine whether the grid purchase price of electricity exceeds the preset price threshold.

[0193] In this step, it is further considered that the cost of purchasing electricity during peak hours under the time-of-use pricing mechanism is usually several times that of the off-peak hours, constituting an important part of the overall operating cost. The energy storage system has the ability to charge during off-peak hours and discharge during peak hours, thus shaving off the peak. The diesel generator can purchase electricity instead of the grid when the electricity price is higher than its generation cost, and the thermal storage tank can store the heat generated during off-peak hours for use during peak hours. Therefore, in order to minimize the cost of purchasing electricity and avoid high-priced electricity purchases during peak hours, this embodiment introduces a triggering mechanism based on an electricity price threshold. When the grid electricity price exceeds a preset threshold, the energy storage unit is controlled to discharge and the diesel generator is started to generate electricity to replace the grid purchase, while reducing the power consumption of the electric heat pump and meeting the steam load demand based on the heat pre-stored in the thermal storage tank.

[0194] Specifically, the system obtains the current grid electricity purchase price from the grid dispatch center or power trading platform through a communication interface, which is the time-of-use price in the aforementioned embodiment. The electricity price is then compared to a preset electricity price threshold. In one embodiment, the preset electricity price threshold is set to twice the average electricity price of the day. The time is immediately determined to be a peak electricity price period. In another embodiment, the preset electricity price threshold can also be directly set to a fixed value, such as 1.2 yuan / kWh. When the current grid purchase price of electricity exceeds this value, a response is triggered. The specific value of the preset electricity price threshold can be flexibly adjusted according to the factory's actual electricity cost affordability and historical electricity price distribution.

[0195] S505. When the preset electricity price threshold is exceeded, the energy storage unit is controlled to discharge and the diesel generator is started to generate electricity to replace the purchase of electricity from the grid.

[0196] In this step, when the current grid electricity price exceeds a preset price threshold, the system first sends a discharge command to the energy storage converter, controlling the energy storage unit to operate in discharge mode and release the stored electrical energy to the microgrid bus to handle the power load that would otherwise need to be purchased from the grid. The discharge power is determined based on the current power shortage and the remaining state of charge of the energy storage unit. Under the premise of meeting the state of charge limit constraints, the maximum allowable discharge power is prioritized to maximize the substitution of high-priced grid electricity.

[0197] Furthermore, if the energy storage unit's discharge power cannot fully cover the power deficit, the system simultaneously sends a start command and a power command to the diesel generator set, initiating diesel power generation so that it, along with the energy storage unit, can share the power deficit originally supplemented by the grid. The diesel generator's output gradually increases to the target value according to its economic output range and ramp rate constraints until the power balance constraint is met. In one embodiment, if the current power deficit is 80kW and the maximum discharge power of the energy storage is 50kW, the diesel generator will supplement the remaining 30kW; in another embodiment, if the electricity price is higher than the marginal cost of diesel generator generation, the system can prioritize increasing the diesel generator's output to its high-efficiency range (e.g., 70%~90% of rated power) to further reduce the unit power generation cost.

[0198] S506, reduce the power consumption of the electric heat pump and meet the steam load demand based on the heat pre-stored in the heat storage tank.

[0199] In this step, simultaneously with the energy storage discharge and diesel generator start-up, the controller sends a command to the heat pump inverter to reduce the input power of the heat pump compressor or inverter drive. This causes the heat pump's power consumption to decrease, freeing up more power capacity for other loads and reducing total power demand. The heat pump's heating power decreases as its power consumption decreases, and this reduced heating capacity is replaced by heat pre-stored in the heat storage tank.

[0200] In practice, the controller reads the current available heat storage capacity, output temperature, and heat loss status of the thermal storage tank, and controls the thermal storage tank to supply heat to the steam generation unit through the heat exchange loop to meet the steam load demand. If the available heat storage capacity of the thermal storage tank can cover the steam demand for that period, the thermal storage tank will release heat to fully compensate for the demand, and the electric heat pump can operate at its lowest power consumption. If the heat storage capacity can only cover part of the steam demand, the power of the electric heat pump will be appropriately reduced to a balanced level, and the remaining heat gap will be borne by the thermal storage tank. If the heat storage capacity is sufficient to cover all the steam demand for that period and there is a surplus, the electric heat pump will maintain its lowest power consumption and the thermal storage tank will be used to supply steam, minimizing the demand for high-priced electricity.

[0201] In this approach, the system triggers alternative power supply on the electricity side based on an electricity price threshold, simultaneously reducing the power consumption of the electric heat pump and utilizing the stored heat in the thermal storage tank to compensate for steam demand. This allows the electricity and heat sides to switch in tandem under high electricity purchase price scenarios, reducing reliance on high-priced grid electricity and maintaining a continuous steam supply. Since the energy storage unit, diesel generator, and thermal storage tank are all coordinated by a unified controller, their output switching can match the current load status, thereby keeping the overall operating cost at a low level and ensuring the stability of power and steam supply for the textile process.

[0202] For example, a large textile printing and dyeing factory obtains a real-time grid purchase price of 1.05 yuan / kWh at 18:00, while the daily average price is 0.5 yuan / kWh. The system determines that 1.05 > 2 × 0.5 = 1, which is valid. The system then enters peak-price response mode: it sends a discharge command to the energy storage converter, which discharges at 0.8MW; it also sends a start command to the diesel generator set, which operates at 0.2MW, with both sharing the power supply. Simultaneously, the controller reduces the power consumption of the electric heat pump to 1.5MW, and the heat storage tank supplies heat to the steam header at 1.0MW to compensate for the steam shortfall caused by the reduced power of the electric heat pump, maintaining a stable steam load of 4.0MW. During this period, the grid purchase power drops to 0.5MW, avoiding full utilization of the high-priced peak electricity at 1.05 yuan / kWh. If the electricity purchase during this period were entirely dependent on the grid, the cost would be 1.05 × 4.0 × 1 = 4200 yuan. In this plan, the photovoltaic power generation is 0.5MW (cost 0), the energy storage discharge is 0.8MW (off-peak charging cost is 0.30 yuan / kWh, and about 0.38 yuan / kWh after considering the charging and discharging efficiency), the diesel power generation is 0.2MW (fuel cost is about 0.65 yuan / kWh), the grid power generation is 0.5MW (1.05 yuan / kWh), and the heat released by the thermal storage tank is low-cost thermal energy pre-produced during off-peak hours. The total cost is about 2950 yuan, saving about 30%.

[0203] This application provides a multi-energy flow collaborative optimization scheduling method. By acquiring the current photovoltaic (PV) output and power load, when the PV output exceeds the sum of the power load and the minimum power consumption of the electric heat pump, the power consumption of the electric heat pump is increased and the heat is stored in a thermal storage tank. When the sum of the PV output, the maximum energy storage charging and discharging power, and the maximum diesel output is still less than the sum of the power load and the minimum power consumption of the electric heat pump, the power consumption of the electric heat pump is reduced, the heat release from the thermal storage tank is increased, and the boiler is started when necessary. This combination of measures achieves active absorption of PV surplus during periods of surplus PV power and coordinated supply guarantee during periods of power deficit. This method maximizes the utilization of PV power generation resources and effectively avoids curtailment. At the same time, during periods of power deficit, the multi-level collaborative strategy of releasing power-side capacity by reducing the power of the electric heat pump, maintaining steam load by releasing heat from the thermal storage tank, and supplementing the boiler as a backup ensures the reliability of the system's power supply and the continuity of steam supply. Meanwhile, by obtaining the current grid electricity price and comparing it with a preset price threshold, the system controls the energy storage unit to discharge and starts a diesel generator to replace grid electricity purchase when the price exceeds the threshold. Simultaneously, it reduces the power consumption of the electric heat pump and utilizes the pre-stored heat in the thermal storage tank to meet steam load demand. This combined approach proactively avoids and replaces high-priced electricity purchases during peak hours. This method fully leverages the peak-shaving and valley-filling capabilities of the energy storage unit, the fuel-powered electricity replacement capability of the diesel generator, and the cross-period heat transfer capability of the thermal storage tank. While ensuring a balance between power supply and demand and stable steam load, it effectively reduces electricity purchase costs during peak hours and improves the system's operational economy.

[0204] Figure 6 This is a schematic diagram of a multi-energy flow collaborative optimization scheduling device provided in an embodiment of this application. Figure 6 As shown, the multi-energy flow collaborative optimization scheduling device 60 includes:

[0205] Module 601 is used to establish a multi-energy flow coupling model that includes power networks, heat networks and carbon emission flows. The multi-energy flow coupling model is used to dynamically describe the power balance relationship and carbon emission factor of each energy node through real-time operating data.

[0206] The scheduling module 602 is used to solve a mixed integer programming problem with the objective function of minimizing the overall operating cost, and to obtain the scheduling plan for the first time period in the future. The overall operating cost includes fuel cost, electricity purchase cost, carbon cost and equipment depreciation cost.

[0207] The scheduling module 602 is also used to collect current actual running data during the execution of the scheduling plan, and use the current actual running data as the initial state to predict and control the scheduling instructions in the future second time period in the scheduling plan through model prediction and rolling correction, the duration of the future second time period is less than the duration of the future first time period;

[0208] The control module 603 is used to adjust at least one of the following when a power fluctuation is detected: the power of the electric heat pump, the heat storage tank charging and discharging power, and the energy demand of the thermal elastic load, in order to respond to the power fluctuation and maintain the stability of the system energy supply.

[0209] In one possible embodiment, the construction module 601 is further configured to:

[0210] The power dynamic balance constraint of the power network is established based on the photovoltaic output, the charging and discharging power of the energy storage unit, the diesel output, the grid switching power and the power consumption of the electric heat pump, and the real-time state of charge of the energy storage unit is updated based on the dynamic equation of the energy storage state of charge.

[0211] The dynamic balance constraint of thermal power in the thermal network is established based on the heating power of electric heat pump, waste heat recovery power, boiler heating power and heat storage tank charging and discharging power, and the real-time total heat storage of the heat storage tank is calculated based on the layered thermodynamic model of the heat storage tank.

[0212] The instantaneous carbon emission intensity of the power load is determined by the weighted average of the carbon emission factors and their output proportions of photovoltaic power generation, energy storage unit charging and discharging, diesel power generation and grid power purchase.

[0213] The instantaneous carbon emission intensity of the steam load is determined by the weighted average of the carbon emission factors of the electric heat pump, waste heat recovery and boiler and their respective heat supply ratios; among them, the carbon emission factor of the electric heat pump is determined by the ratio of the instantaneous carbon emission intensity to the energy efficiency coefficient of the electric heat pump.

[0214] The total system carbon emissions of the carbon emission flow are calculated based on the cumulative values ​​of the product of diesel power generation and diesel carbon emission factor, the product of grid exchange power and grid real-time carbon emission factor, and the product of boiler heating power and boiler carbon emission factor within the dispatch cycle.

[0215] In one possible embodiment, before solving the mixed-integer programming problem with the objective function of minimizing the overall operating cost, the construction module 601 is further configured to:

[0216] Obtain the delay time, acceptable deviation of steam supply, and priority of each heat-consuming process;

[0217] The thermoelasticity coefficients of each heat-consuming process are determined based on the delay time, acceptable deviation, and priority. The thermoelasticity coefficients are used to characterize the tolerance of the corresponding heat-consuming process to fluctuations in steam supply.

[0218] In one possible embodiment, the scheduling module 602 is further configured to:

[0219] Using the first future time period as the scheduling cycle, the scheduling cycle is discretized into multiple scheduling periods according to a preset time step. The objective function is to minimize the sum of fuel cost, electricity purchase cost, carbon cost, energy storage unit life loss cost and thermal storage tank life loss cost. A day-ahead scheduling optimization model is then established.

[0220] The constraints of the day-ahead dispatch optimization model are determined, including dynamic balance constraints of power output, dynamic balance constraints of thermal power output, operational constraints of energy storage units, operational constraints of diesel generators, operational constraints of thermal storage tanks, and thermoelastic constraints of steam load.

[0221] A mixed-integer programming solver is used to solve the day-ahead scheduling optimization model to obtain the scheduling plan for each device in the first future time period.

[0222] In one possible embodiment, the scheduling module 602 is further configured to:

[0223] Using the second future time period as the prediction time domain, the prediction time domain is divided into multiple control intervals according to the preset control cycle;

[0224] Using the current actual operating data as the initial state, and with the optimization objective of minimizing the sum of fuel cost, electricity purchase cost, carbon emission cost, penalty term for changes in power dispatch instructions, and penalty term for changes in heat dispatch instructions, an intraday rolling optimization model is established. The penalty term for changes in power dispatch instructions is determined based on the deviation of power dispatch instructions between the current time and the previous control interval, and the penalty term for changes in heat dispatch instructions is determined based on the deviation of heat dispatch instructions between the current time and the previous control interval.

[0225] Based on the constraints of the day-ahead scheduling optimization model, the constraints of the intraday rolling optimization model are determined.

[0226] At the arrival of each control cycle, a mixed integer programming solver is used to solve the intraday rolling optimization model, and the scheduling instructions obtained in the current control interval are executed as the corrected scheduling instructions. The rolling solution process is repeated at the arrival of the next control cycle.

[0227] In one possible embodiment, the control module 603 is further configured to:

[0228] When a power shortage is detected, the power consumption of the electric heat pump is reduced and the heat release power of the heat storage tank is increased to compensate for the steam supply gap caused by the reduced power of the electric heat pump.

[0229] When a steam power deficit is detected, the heat release power of the thermal storage tank is increased first to respond to the steam power deficit;

[0230] If there is still a steam supply gap after the heat storage tank has reached its maximum allowable heat release power, the steam supply of the corresponding heat-using processes will be reduced in order of decreasing thermoelasticity coefficients to prioritize the steam supply of heat-using processes with lower thermoelasticity coefficients.

[0231] In one possible embodiment, the control module 603 is further configured to:

[0232] Obtain the current photovoltaic output and power load;

[0233] When the photovoltaic output exceeds the sum of the power load and the minimum power consumption of the electric heat pump, the power consumption of the electric heat pump is increased to convert the surplus electrical energy into heat energy and store it in the heat storage tank.

[0234] When the sum of photovoltaic output, maximum energy storage unit charging and discharging power, and maximum diesel output is less than the sum of electrical load and minimum electric heat pump power consumption, the power consumption of the electric heat pump is reduced and the heat release power of the heat storage tank is increased to meet the steam load. When the heat supply from the heat storage tank is insufficient, the boiler is started to supplement the steam supply.

[0235] In one possible embodiment, the control module 603 is further configured to:

[0236] Obtain the current grid purchase price of electricity and determine whether the grid purchase price of electricity exceeds the preset price threshold;

[0237] When the preset electricity price threshold is exceeded, the energy storage unit is controlled to discharge and the diesel generator is started to generate electricity to replace the purchase of electricity from the grid;

[0238] Reduce the power consumption of the electric heat pump and meet the steam load demand based on the heat stored in the heat storage tank.

[0239] This embodiment provides a multi-energy flow collaborative optimization scheduling device that can execute the multi-energy flow collaborative optimization scheduling method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0240] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 7 As shown, the electronic device 70 includes a processor 701 and a memory 702 communicatively connected to the processor 701. Optionally, the electronic device 70 also includes a communication component 703. The processor 701, memory 702, and communication component 703 are connected via a bus 704.

[0241] Memory 702 stores instructions executed by the computer;

[0242] The processor 701 executes computer execution instructions stored in the memory 702 to implement the multi-energy flow cooperative optimization scheduling method as described above.

[0243] At least one processor 701 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0244] Optionally, in specific implementations, the processor 701 and memory 702 are implemented independently. In this case, the processor 701 and memory 702 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc., but this does not imply that there is only one bus or one type of bus.

[0245] Optionally, in a specific implementation, if the processor 701 and the memory 702 are integrated on a single chip, the processor 701 and the memory 702 can communicate through an internal interface.

[0246] This application also provides a computer storage medium storing computer execution instructions, which, when executed by a processor, implements the aforementioned multi-energy flow cooperative optimization scheduling method.

[0247] The aforementioned computer-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The computer-readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0248] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the control device of the present application.

[0249] This unit division is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0250] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0251] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0252] If this function is implemented as a software functional unit and used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods indicated in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0253] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0254] The technical solutions of this application have been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it is readily understood by those skilled in the art that the scope of protection of this application is obviously not limited to these specific embodiments. The above embodiments are only used to illustrate the technical solutions of this application and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A multi-energy flow cooperative optimization scheduling method, characterized in that, The method includes: A multi-energy flow coupling model is established, which includes power networks, heat networks, and carbon emission flows. The multi-energy flow coupling model is used to dynamically describe the power balance relationship and carbon emission factor of each energy node through real-time operating data. The mixed-integer programming problem is solved with the objective function of minimizing the overall operating cost to obtain the scheduling plan for the first time period in the future. The overall operating cost includes fuel cost, electricity purchase cost, carbon cost, and equipment depreciation cost. During the execution of the scheduling plan, current actual operating data is collected, and the scheduling instructions in the future second time period in the scheduling plan are rolled and corrected through model prediction control using the current actual operating data as the initial state. The duration of the future second time period is less than the duration of the future first time period. In the event of a detected power fluctuation, at least one of the following is adjusted in a coordinated manner: the power of the electric heat pump, the heat storage tank charging and discharging power, and the energy demand of the thermal elastic load, in order to respond to the power fluctuation and maintain a stable system energy supply. The establishment of a multi-energy flow coupling model that includes power networks, thermal networks, and carbon emission flows includes: The power dynamic balance constraint of the power network is established based on the photovoltaic output, the charging and discharging power of the energy storage unit, the diesel output, the grid switching power, and the power consumption of the electric heat pump, and the real-time state of charge of the energy storage unit is updated based on the dynamic equation of the energy storage state of charge. The dynamic balance constraint of the thermal power of the thermal network is established based on the heating power of the electric heat pump, the waste heat recovery power, the heating power of the boiler, and the charging and discharging power of the thermal storage tank. The real-time total heat storage of the thermal storage tank is calculated based on the layered thermodynamic model of the thermal storage tank. The instantaneous carbon emission intensity of the power load is determined by the weighted average of the carbon emission factors and their output proportions of photovoltaic power generation, energy storage unit charging and discharging, diesel power generation and grid power purchase. The instantaneous carbon emission intensity of the steam load is determined by the weighted average of the carbon emission factors of the electric heat pump, waste heat recovery, and boiler, and their respective heat supply ratios; wherein, the carbon emission factor of the electric heat pump is determined by the ratio of the instantaneous carbon emission intensity to the energy efficiency coefficient of the electric heat pump. The total system carbon emissions of the carbon emission stream are calculated based on the cumulative values ​​of the product of diesel power generation and diesel carbon emission factor, the product of grid exchange power and grid real-time carbon emission factor, and the product of boiler heating power and boiler carbon emission factor within the scheduling cycle.

2. The method according to claim 1, characterized in that, Before solving the mixed-integer programming problem with the objective function of minimizing the overall operating cost, the method further includes: Obtain the delay time, acceptable deviation of steam supply, and priority of each heat-consuming process; The thermoelasticity coefficient of each heat-using process is determined based on the delay time, the acceptable deviation, and the priority, wherein the thermoelasticity coefficient is used to characterize the tolerance of the corresponding heat-using process to fluctuations in steam supply.

3. The method according to claim 2, characterized in that, The method of solving a mixed-integer programming problem with the objective function of minimizing the overall operating cost to obtain the scheduling plan for the first time period includes: Using the first future time period as the scheduling cycle, the scheduling cycle is discretized into multiple scheduling time periods according to a preset time step. The objective function is to minimize the sum of fuel cost, electricity purchase cost, carbon cost, energy storage unit life loss cost and thermal storage tank life loss cost. A day-ahead scheduling optimization model is then established. The constraints of the day-ahead scheduling optimization model are determined, wherein the constraints include dynamic balance constraints of power output, dynamic balance constraints of thermal power output, operation constraints of energy storage units, operation constraints of diesel generators, operation constraints of thermal storage tanks, and thermoelastic constraints of steam load. The day-ahead scheduling optimization model is solved using a mixed integer programming solver to obtain the scheduling plan for each device in the first future time period.

4. The method according to claim 3, characterized in that, The step of using the current actual operating data as the initial state and rollingly correcting the scheduling instructions in the future second time period of the scheduling plan through model prediction control includes: Using the second future time period as the prediction time domain, the prediction time domain is divided into multiple control intervals according to a preset control cycle; Using the current actual operating data as the initial state, and with the optimization objective of minimizing the sum of fuel cost, electricity purchase cost, power dispatch command change penalty term and thermal dispatch command change penalty term, an intraday rolling optimization model is established. The power dispatch command change penalty term is determined based on the power dispatch command deviation between the current time and the previous control interval, and the thermal dispatch command change penalty term is determined based on the thermal dispatch command deviation between the current time and the previous control interval. Based on the constraints of the day-ahead scheduling optimization model, the constraints of the intraday rolling optimization model are determined. At the arrival of each control cycle, a mixed integer programming solver is used to solve the intraday rolling optimization model, and the scheduling instructions obtained in the current control interval are executed as the corrected scheduling instructions. The rolling solution process is repeated at the arrival of the next control cycle.

5. The method according to claim 4, characterized in that, The linkage regulation of at least one of the following: the power of the electric heat pump, the heat storage tank charging and discharging power, and the energy demand of the thermal elastic load, including: When a power shortage is detected, the power consumption of the electric heat pump is reduced and the heat release power of the heat storage tank is increased to compensate for the steam supply gap caused by the reduced power of the electric heat pump. When a steam power deficit is detected, the heat release power of the thermal storage tank is increased first in response to the steam power deficit; If there is still a steam supply gap after the heat storage tank has reached its maximum allowable heat release power, the steam supply of the corresponding heat-using processes will be reduced in order of decreasing thermoelasticity coefficients to prioritize the steam supply of heat-using processes with lower thermoelasticity coefficients.

6. The method according to any one of claims 2-5, characterized in that, The method further includes: Obtain the current photovoltaic output and power load; When the photovoltaic output is greater than the sum of the power load and the minimum power consumption of the electric heat pump, the power consumption of the electric heat pump is increased to convert the surplus electrical energy into heat energy and store it in the heat storage tank. When the sum of the photovoltaic output, the maximum energy storage unit charging and discharging power, and the maximum diesel output is less than the sum of the power load and the minimum power consumption of the electric heat pump, the power consumption of the electric heat pump is reduced and the heat release power of the heat storage tank is increased to meet the steam load. When the heat supply of the heat storage tank is insufficient, the boiler is started to supplement the steam supply.

7. The method according to claim 6, characterized in that, The method further includes: Obtain the current grid electricity purchase price and determine whether the grid electricity purchase price exceeds a preset electricity price threshold; When the preset electricity price threshold is exceeded, the energy storage unit is controlled to discharge and the diesel generator is started to generate electricity to replace the purchase of electricity from the grid; Reduce the power consumption of the electric heat pump and meet the steam load demand based on the heat stored in the heat storage tank.

8. A multi-energy flow collaborative optimization scheduling device, characterized in that, include: The module is used to establish a multi-energy flow coupling model that includes power networks, thermal networks, and carbon emission flows. The multi-energy flow coupling model is used to dynamically describe the power balance relationship and carbon emission factor of each energy node through real-time operating data. The scheduling module is used to solve a mixed integer programming problem with the objective function of minimizing the overall operating cost, and to obtain the scheduling plan for the first time period in the future. The overall operating cost includes fuel cost, electricity purchase cost, carbon cost, and equipment depreciation cost. The scheduling module is also used to collect current actual operating data during the execution of the scheduling plan, and use the current actual operating data as the initial state to continuously correct the scheduling instructions in the future second time period in the scheduling plan through model prediction control, wherein the duration of the future second time period is less than the duration of the future first time period; The control module is used to adjust at least one of the following when a power fluctuation is detected: the power of the electric heat pump, the heat storage tank charging and discharging power, and the energy demand of the thermal elastic load, in order to respond to the power fluctuation and maintain the stability of the system's energy supply. The establishment of a multi-energy flow coupling model that includes power networks, thermal networks, and carbon emission flows includes: The power dynamic balance constraint of the power network is established based on the photovoltaic output, the charging and discharging power of the energy storage unit, the diesel output, the grid switching power, and the power consumption of the electric heat pump, and the real-time state of charge of the energy storage unit is updated based on the dynamic equation of the energy storage state of charge. The dynamic balance constraint of the thermal power of the thermal network is established based on the heating power of the electric heat pump, the waste heat recovery power, the heating power of the boiler, and the charging and discharging power of the thermal storage tank. The real-time total heat storage of the thermal storage tank is calculated based on the layered thermodynamic model of the thermal storage tank. The instantaneous carbon emission intensity of the power load is determined by the weighted average of the carbon emission factors and their output proportions of photovoltaic power generation, energy storage unit charging and discharging, diesel power generation and grid power purchase. The instantaneous carbon emission intensity of the steam load is determined by the weighted average of the carbon emission factors of the electric heat pump, waste heat recovery, and boiler, and their respective heat supply ratios; wherein, the carbon emission factor of the electric heat pump is determined by the ratio of the instantaneous carbon emission intensity to the energy efficiency coefficient of the electric heat pump. The total system carbon emissions of the carbon emission stream are calculated based on the cumulative values ​​of the product of diesel power generation and diesel carbon emission factor, the product of grid exchange power and grid real-time carbon emission factor, and the product of boiler heating power and boiler carbon emission factor within the scheduling cycle.

9. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the multi-energy flow cooperative optimization scheduling method as described in any one of claims 1 to 7.

Images