Park comprehensive energy optimization scheduling system based on multi-energy flow cooperation
By using a clear-sky model and LSTM neural network to predict photovoltaic output and analyze load time fluctuations, combined with an optimized scheduling model based on the substitution elasticity coefficient, the supply-side uncertainty caused by photovoltaic output fluctuations and natural gas pressure changes was solved, achieving efficient, stable, and economical scheduling of the park's energy system.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- TIANJIN ANJIE PUBLIC FACILITIES SERVICE CO LTD
- Filing Date
- 2026-04-17
- Publication Date
- 2026-05-15
AI Technical Summary
Existing technologies are insufficient to effectively address the supply-side uncertainties of the park's energy system caused by fluctuations in photovoltaic output and changes in natural gas pressure. They lack high-precision forecasting mechanisms and load-side elasticity quantification methods, resulting in high operating costs and insufficient safety and stability.
A photovoltaic output prediction model based on clear sky model and LSTM neural network is adopted. Combined with the optimal scheduling model of load time fluctuation and substitution elasticity coefficient, the model is constructed to minimize the total cost, accurately predict photovoltaic output and natural gas supply, identify transferable and substitutable loads, and optimize energy allocation.
It enables more accurate photovoltaic output forecasting and natural gas supply management, reduces operating costs, improves system flexibility and safety margin, and enhances the stability and economic benefits of the park's energy system.
Smart Images
Figure CN122047657A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of energy dispatching technology, specifically to a comprehensive energy optimization dispatching system for industrial parks based on multi-energy flow collaboration. Background Technology
[0002] As a hub for energy consumption, the industrial park's energy system is transforming from a traditional, single, rigid, and discrete model to a comprehensive, flexible, and collaborative one. The park's Integrated Energy System (IES) integrates multiple energy forms, including electricity, gas, heat, and cooling, enabling multi-energy complementarity and efficient energy use. However, the complexity of this multi-energy coupling also brings unprecedented challenges: photovoltaic output is highly random and fluctuating due to weather conditions, while the natural gas pipeline network's transmission capacity is strictly constrained by dynamic pressure changes, significantly increasing supply-side uncertainty.
[0003] Existing methods often lag in responding to supply-side changes such as intermittent photovoltaic power generation and natural gas pressure fluctuations, relying primarily on passive measures like load shedding rather than proactive adjustments to ensure system safety margins. The root cause lies in the lack of high-precision source-side forecasting mechanisms and refined load-side elasticity quantification methods: traditional forecasting models struggle to isolate the nonlinear impact of meteorological factors on photovoltaic output, leading to significant scheduling deviations; simultaneously, existing scheduling strategies largely treat load as rigid demand, neglecting the substitution potential between electricity and gas loads and the dynamic limitations of pipeline pressure on gas supply capacity, failing to tap into demand-side adjustment resources. This results in the system being unable to anticipate risks when facing source-side fluctuations, nor can it proactively optimize using the spatiotemporal transfer characteristics and multi-energy substitution elasticity of loads, ultimately leading to high operating costs and insufficient safety and stability. Summary of the Invention
[0004] In view of the above, it is necessary to provide a comprehensive energy optimization and scheduling system for industrial parks based on multi-energy flow coordination to solve the above problems.
[0005] One embodiment of this application provides a comprehensive energy optimization and scheduling system for industrial parks based on multi-energy flow coordination, the system comprising:
[0006] The integrated energy dispatch data preparation module of the park obtains the photovoltaic output benchmark value of the park based on the clear sky model, and compares it with the actual photovoltaic output of the park to obtain the output coefficient; it drives the LSTM neural network to predict the future output coefficient based on the meteorological data of the park, and combines it with the photovoltaic output benchmark value to obtain the final photovoltaic predicted output. The park energy load judgment module is used to analyze the electricity consumption periods of each load device in the park during the statistical period. Based on the overlap duration and number of overlap periods, the load time fluctuation of each load device is determined. Based on the theoretical maximum flow rate of the park's gas pipeline at the current pressure, combined with the predicted total gas load of the park in the future scheduling period, the adequacy of natural gas supply is determined. The similarity between the natural gas supply adequacy and the load power of the load devices in the historical statistical period is analyzed. Combined with the load time fluctuation, the substitution elasticity coefficient of the load devices is obtained, and the load type of the load devices is determined. The integrated energy dispatch module of the park is used to construct an optimal dispatch model with the goal of minimizing total cost. Based on the power balance of the supply and demand sides of the park, as well as the start-stop status and total power demand of transferable loads during the dispatch cycle, the constraints of the optimal dispatch model are obtained, and the optimal energy dispatch scheme is obtained by solving the problem. The determination of the load time fluctuation of each load device specifically involves: Obtain the daily electricity consumption period of each load device, overlap the daily electricity consumption periods of different dates within the set statistical period, and take the overlapping time intervals covered by the electricity consumption periods of each date as the characteristic time period of each load device. The load time fluctuation of each load device is obtained by summing the reciprocals of the number of electricity consumption periods corresponding to each characteristic period. The obtained replacement elasticity coefficient of the load equipment is specifically as follows: The correlation between load power and natural gas supply adequacy over a historical statistical period is obtained and denoted as p; Alternating elasticity coefficient The specific formula is: Where max() represents the maximum value function, and T represents the normalized result of the load time fluctuation of the load equipment.
[0007] Preferably, the specific steps for obtaining the output coefficient include: When the photovoltaic power output benchmark value is less than the preset clear sky benchmark threshold, the power output coefficient k is set to 0; when the photovoltaic power output benchmark value is greater than or equal to the preset clear sky benchmark threshold, the ratio of the actual photovoltaic power output to the photovoltaic power output benchmark value is used as the power output coefficient.
[0008] Preferably, the final photovoltaic power output is specifically the product of the photovoltaic power output benchmark value and the predicted future power output coefficient.
[0009] Preferably, the process of determining the adequacy of natural gas supply is as follows: The theoretical maximum flow rate at the current inlet pressure of the gas pipeline network access point in the park is used as the upper limit of the gas supply capacity of the gas pipeline network in the corresponding statistical period; the total gas load forecast value is obtained by using a prediction model based on the historical gas consumption of the gas pipeline network in the park. The ratio of the upper limit of the gas supply capacity of the park's gas pipeline network to the predicted value of the total gas load is used as the sufficiency of natural gas supply.
[0010] Preferably, determining the load type of the load equipment specifically involves: For each load device, if the normalized result of the load time fluctuation is greater than or equal to the first characteristic threshold, the load type of the load device is a transferable load; if the normalized result of the load time fluctuation is less than the first characteristic threshold, the correlation is less than 0, and the substitution elasticity coefficient is greater than or equal to the second characteristic threshold, the load type of the load device is a substituteable load; otherwise, it is a rigid load. Specifically, the first characteristic threshold is the average value of the load time fluctuation of all load devices, and the second characteristic threshold is the average value of the substitution elasticity coefficient of all load devices.
[0011] Preferably, the objective function of the optimized scheduling model is: ;in, This represents the total price of electricity purchased from the grid during time period t. The cost of purchasing gas from the gas network during time period t. To compensate for the cost of demand response, Represents the function to be minimized. The objective function value is represented by s, which represents the number of time periods. The time periods are determined by dividing the preset scheduling cycle equally.
[0012] Preferably, the constraints for obtaining the optimized scheduling model are specifically as follows: With the balance between the total power supply and the total power demand as a constraint, by setting the gas-to-electricity equivalent coefficient, for each replaceable load, the sum of the electricity consumed in each time period and the equivalent gas volume is equal to the original total energy demand forecast value for that time period. Ensure that the total power consumption of each transferable load throughout the entire scheduling cycle is equal to the total power demand set by the user, and that it only operates within the time intervals allowed by the user.
[0013] Preferably, the energy dispatch scheme includes: electricity purchase plans from the power grid for each time period within a preset dispatch cycle, gas purchase plans from the gas grid for each time period within a preset dispatch cycle, electricity / gas allocation instructions for each replaceable load, and start / stop time instructions for each transferable load.
[0014] This application has at least the following beneficial effects: The data preparation module in this application can effectively identify performance deviations in photovoltaic power generation by comparing benchmark values with actual output, providing a more accurate output coefficient. By using LSTM neural networks to predict future photovoltaic output, it is possible to grasp photovoltaic power generation capacity in advance, which helps to formulate reasonable scheduling plans. Finally, more accurate photovoltaic output prediction provides a scientific basis for energy management and scheduling in the park, which helps to balance supply and demand.
[0015] The energy load assessment module identifies peak and off-peak electricity consumption periods by statistically analyzing the electricity usage times of various load devices, which helps adjust dispatching strategies. By constructing load time fluctuations, it can accurately identify transferable loads, avoiding treating such flexible loads as rigid loads. It assesses the adequacy of natural gas supply to ensure that user demand can be met during periods of high demand, thereby avoiding energy waste or equipment damage due to insufficient gas supply. By calculating the substitution elasticity coefficient of load devices, it can better understand the response capabilities of each load device under different conditions, quantify the comprehensive potential of electrical load devices (such as electric water heaters) to switch to natural gas energy when gas supply is abundant, and thus accurately distinguish between substitutable loads and rigid loads.
[0016] The integrated energy dispatch module effectively reduces the park's operating costs and improves economic efficiency by minimizing the total cost of dispatch. It also ensures power balance between the supply and demand sides to avoid energy waste or shortages and improve the stability of park operations. By comprehensively considering the start-up and shutdown status of transferable loads, the park can quickly respond to sudden changes in electricity demand, improving the system's flexibility and adaptability.
[0017] By optimizing the electricity / gas allocation ratio for alternative loads through the coordinated operation of the aforementioned modules, the park can encourage gas substitution for electricity when gas prices are low or gas supply is abundant, and vice versa, thereby reducing overall energy costs. Simultaneously, by smoothing the load curve, the park improves the absorption rate of distributed energy sources such as photovoltaics and combined heat and power (CHP), alleviating power supply pressure during peak hours, avoiding equipment failures or power outages due to overload, and enhancing the overall safety margin and operational stability of the park's energy system to achieve more efficient energy dispatch. This integrated dispatch scheme not only meets the park's current energy needs but also has strong foresight, preparing for potential future changes in energy demand, thus ensuring the park's sustainable development. Attached Figure Description
[0018] Figure 1 A block diagram of a multi-energy flow collaborative integrated energy optimization and dispatching system for industrial parks, provided for this application; Figure 2 The following is a flowchart of the integrated energy optimization and scheduling process for the park provided in this application. Detailed Implementation
[0019] In the description of the embodiments in this application, the words "exemplary," "or," and "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary," "or," and "for example" is intended to present the relevant concepts in a specific manner.
[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in this application's specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.
[0021] It should also be noted that the terms "first" and "second" in this application and its accompanying drawings are used to distinguish similar objects, rather than to describe a specific order or sequence. The methods disclosed in the embodiments of this application or the methods shown in the flowcharts include one or more steps for implementing the method. Without departing from the scope of protection of this application, the execution order of multiple steps can be interchanged, and some steps can also be deleted.
[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.
[0023] The following description, in conjunction with the accompanying drawings, details the specific scheme of the integrated energy optimization and scheduling system for industrial parks based on multi-energy flow collaboration provided in this application.
[0024] Please see Figure 1 The diagram illustrates a block diagram of a multi-energy flow collaborative integrated energy optimization and scheduling system for industrial parks, according to an embodiment of this application. The system includes: an integrated energy scheduling data preparation module, an energy load judgment module, and an integrated energy scheduling module. Please refer to... Figure 2 The diagram illustrates a specific flowchart of a park integrated energy optimization scheduling provided in one embodiment of this application.
[0025] This application first proposes a multi-energy flow collaborative integrated energy optimization and scheduling system for industrial parks, applied in the field of energy scheduling technology. The system includes: The integrated energy dispatch data preparation module of the park obtains the photovoltaic output benchmark value of the park based on the clear sky model, and compares it with the actual photovoltaic output of the park to obtain the output coefficient; it drives the LSTM neural network to predict the future output coefficient based on the meteorological data of the park, and combines it with the photovoltaic output benchmark value to obtain the final photovoltaic predicted output.
[0026] Historical load power curves, pipeline operation data, meteorological information, real-time electricity / gas prices, and equipment physical parameters are collected through the park's EMS, gas pipeline SCADA, meteorological API, and equipment ledgers.
[0027] Specifically, the system obtains data such as inlet and outlet pressures of the pressure regulating station, real-time flow rate, gas temperature in the pipeline network, pressure-flow characteristic curves of the pressure regulating station, and historical gas consumption through the gas pipeline network SCADA system and equipment ledger. It collects total demand-side power and power supply-side power through smart meters and battery management systems. Total demand-side power includes total load power, energy storage charging power, and heat pump power; power supply-side power includes combined heat and power unit power, actual photovoltaic output, and energy storage discharge power. It collects ambient temperature, total irradiance, diffuse irradiance, relative humidity, wind speed, real-time electricity price, and real-time natural gas price through meteorological station APIs and trading interfaces. Finally, it obtains the maximum electrical power, maximum gas power, gas-to-electricity equivalence coefficient, and rated power of transferred loads from the equipment ledger database.
[0028] Furthermore, a combined prediction method of clear-sky model and LSTM is adopted. By combining the clear-sky model with the park's latitude and longitude, timestamp, photovoltaic panel tilt angle, and azimuth angle, the baseline value of photovoltaic output under theoretical clear-sky conditions is calculated. Specifically, the clear-sky model used in this embodiment is the Solargis model, which is a publicly available model and will not be described further in this application. A preset clear-sky baseline threshold is set; in this embodiment, the clear-sky baseline threshold is taken as... When the photovoltaic (PV) power output benchmark value is less than the preset clear-sky benchmark threshold, the power output coefficient k is set to 0; when the PV power output benchmark value is greater than or equal to the preset clear-sky benchmark threshold, the ratio of the actual PV power output to the PV power output benchmark value is used as the power output coefficient k. Min-Max normalization is performed on the total irradiance, ambient temperature, relative humidity, wind speed, and historical k-value sequences, mapping the data to the [0,1] interval. The normalized data is used as the input to the LSTM neural network model, outputting a preset number of predicted k-values for the future scheduling period (specifically, one every 15 minutes in the next hour, i.e., a preset number of 4), with the MAE loss function used. The final predicted PV power output is obtained by multiplying the PV power output benchmark value by the predicted k-values.
[0029] The park energy load judgment module analyzes the electricity consumption periods of each load device within the park during the statistical period. Based on the overlap duration and number of overlap periods, it determines the load time fluctuation of each load device. Based on the theoretical maximum flow rate of the park's gas pipeline at the current pressure, combined with the predicted total gas load of the park during future scheduling periods, it determines the adequacy of natural gas supply. It analyzes the similarity between the natural gas supply adequacy and the load power of the load devices within the historical statistical period, and combined with the load time fluctuation, obtains the substitution elasticity coefficient of the load devices to determine the load type of the load devices.
[0030] Adjustable loads refer to users in an energy system who can flexibly adjust their energy consumption patterns or usage methods based on demand signals. These adjustable load resources can be classified into different types based on their energy consumption characteristics, usage intensity, and responsiveness.
[0031] In energy systems, adjustable load resources optimize energy consumption strategies by participating in demand response, thereby significantly improving system flexibility and responsiveness. This adjustment helps balance energy supply and demand within the park, while simultaneously achieving the goals of energy cost savings and efficiency improvements. In multi-energy systems within a park, load demand response is mainly divided into two categories: transferable loads and substitute loads.
[0032] Transferable loads refer to loads whose energy consumption periods can be flexibly adjusted within a certain time range while maintaining a constant total energy consumption, such as electric vehicles and water heaters. Substitutable loads, on the other hand, refer to loads with fixed energy consumption periods, but which can be adapted to different energy sources based on changes in energy prices or supply conditions, thus changing the energy form, such as electric / gas water heaters.
[0033] Based on the above description, this application analyzes the load demand response type of the load devices connected to the park's energy management system according to the load demand response characteristics. Taking any connected load device Q as an example, the historical daily load power curve of the load device is obtained through the historical records of the park's energy management system. The horizontal axis of the load power curve is time, and the vertical axis is the power of the load device. For substitutable loads, their load curves usually show relatively regular energy demand during the same period (such as morning and evening peak electricity consumption periods), and the duration is relatively short. For transferable loads, their load curves usually show long periods of flat power, but the total electricity consumption is constant, and the time of occurrence is not fixed and the duration fluctuates (e.g., the charging period for electric vehicles may be any time of day, and the length of each charging period may also be inconsistent). Based on this characteristic, the daily power consumption periods of the load equipment are obtained. The average power value of all moments within the past set statistical period (30 days in this embodiment) of the load equipment is used as the feature threshold. On a daily basis, all consecutive moments with power values greater than or equal to the feature threshold are merged to obtain at least one power consumption period on the relative time axis of the 24-hour period of each day. There is a corresponding power consumption period every day in the set statistical period of the load equipment. The daily power consumption periods of different dates within the set statistical period are overlapped and calculated. The overlapping time interval covered by the power consumption periods of each date is used as the feature period of the load equipment. It should be understood that each feature period is obtained by overlapping the power consumption periods of multiple specific historical dates. The load time fluctuation of load equipment is constructed by analyzing the distribution of characteristic time periods of load equipment. The more characteristic time periods there are and the fewer the number of electricity consumption periods corresponding to the characteristic time periods (for example, if the characteristic time period T is the intersection of 10 electricity consumption periods, then the number is 10), the more unstable the energy consumption period of the load equipment is, and the more likely the load equipment is to be a transferable load. Therefore, this application uses the sum of the reciprocals of the number of electricity consumption periods corresponding to each characteristic time period (since there must be a corresponding electricity consumption period for each characteristic time period, the denominator does not have a zero phenomenon) as the load time fluctuation of the load equipment.
[0034] While load fluctuation of load equipment can effectively determine whether its load is transferable, it cannot effectively determine whether its load is replaceable. Therefore, this application further obtains the gas (natural gas) energy supply status in the park's energy management system. Through the park's gas pipeline network SCADA system, it collects operational data from the access points, including inlet pressure, outlet pressure, real-time flow rate, and temperature. Based on the performance curve of the pressure regulating station equipment (pressure-flow characteristic curve provided by the manufacturer), combined with the current inlet pressure, the theoretical maximum flow rate under that pressure is used as the upper limit of the gas network energy supply for the corresponding statistical period. By acquiring multi-dimensional features such as historical gas consumption, meteorological temperature, and holidays from the SCADA system of the park's gas pipeline network, and using the ARIMA-ConvLSTM combined prediction model, a short-term forecast of the park's total gas demand for the future scheduling period is made, resulting in the predicted total gas load for the future scheduling period. The ARIMA-ConvLSTM combined prediction model is an existing publicly available model, which will not be elaborated upon in this application. This allows for the acquisition of natural gas supply adequacy, specifically: natural gas supply adequacy is the ratio of the upper limit of the park's gas pipeline network's energy supply to the predicted total gas load. A larger value indicates a more abundant gas supply, potentially leading to lower gas prices or encouragement of gas consumption; a smaller value indicates a tight gas supply, requiring the activation of demand response or energy switching.
[0035] The above steps completed the acquisition of natural gas supply adequacy. The correlation between the load curves of the load equipment and the natural gas supply adequacy curve was calculated. Specifically, the natural gas supply adequacy sequence within a historical statistical period (the past 30 days in this embodiment) was extracted according to a preset time resolution (15 minutes). Each value in this sequence represents the ratio of the historical upper limit of the gas network's energy supply to the actual historical total gas load at the corresponding moment. The load power sequence of the load equipment within this historical statistical period was extracted, and the corresponding points of the load power sequence and the adequacy sequence were matched and aligned using timestamps. In this application, the Pearson correlation coefficient between the two sequences is used as the correlation, where a positive Pearson correlation coefficient indicates a positive correlation, and a negative Pearson correlation coefficient indicates a negative correlation.
[0036] Furthermore, based on the response intensity of load equipment to changes in natural gas energy supply and the relative regularity of its energy consumption periods, for equipment in the equipment ledger with dual gas and electricity power supply interfaces, a comprehensive assessment is made as to whether its load is a substitutable load. This application constructs a substitution elasticity coefficient for load equipment, which measures the comprehensive potential of load equipment currently consuming electricity to switch to natural gas energy when the natural gas energy supply is abundant. The specific formula for the substitution elasticity coefficient is as follows: Where p represents the correlation between the load curve of the load equipment and the natural gas supply adequacy curve. In this embodiment, the Pearson correlation coefficient is used for calculation. If the standard deviation of the natural gas supply adequacy sequence or the load power sequence is 0, the correlation p is directly assigned a value of 0; max() represents the maximum value function. T represents the normalized result of the load time fluctuation of the load equipment (the normalization method adopts the maximum-minimum normalization method, and the normalization range is to normalize the fluctuation of all load equipment. If the maximum value of the load time fluctuation of all load equipment is equal to the minimum value, the normalization result is directly taken as a fixed reference value of 0.5).
[0037] The above steps have completed the acquisition of the load time fluctuation, the correlation between electrical energy and natural gas energy, and the substitution elasticity coefficient for the load equipment. This allows for the determination of the load type of the load equipment. The first step is to determine if T is greater than or equal to the first characteristic threshold; otherwise, proceed to the second step. The second step determines the correlation between the power of the load equipment (currently electrical energy) and the natural gas energy. If the correlation is less than 0, there is substitution potential. The third step determines if the correlation is greater than or equal to 0, there is no substitution potential, and the load equipment is a rigid load. The third step uses the substitution elasticity coefficient. If the substitution elasticity coefficient is greater than or equal to the second characteristic threshold, the load equipment is a substitutable load; otherwise, it is a rigid load. The first characteristic threshold is the average load time fluctuation of all load equipment, and the second characteristic threshold is the average substitution elasticity coefficient of all load equipment.
[0038] The integrated energy dispatch module of the park constructs an optimized dispatch model with the goal of minimizing total cost. Based on the power balance of the supply and demand sides of the park, as well as the start-stop status and total power demand of transferable loads during the dispatch cycle, the constraints of the optimized dispatch model are obtained, and the optimal energy dispatch scheme is obtained by solving the problem.
[0039] For the park's energy management system, it is necessary to obtain the total power consumed on the demand side, which includes the total electrical power consumed by all load equipment, the charging power of the battery energy storage system, and the electrical power consumed by the heat pump (the total electrical power consumed by the load equipment can be divided into three parts: the predicted value of the base load, the actual power of the transferable load, and the actual electrical power of the substitute load). It is also necessary to obtain the total power generated on the supply side, which includes the electrical power generated by the gas-fired combined heat and power unit, the actual output power of the photovoltaic power generation system, and the discharge power of the battery energy storage system.
[0040] If the total power supply is less than the total power demand, the park's energy management system needs to purchase the corresponding electricity or natural gas from the external power grid. When the total power supply is greater than or equal to the total power demand, there is no need to purchase the corresponding electricity or natural gas from the external power grid. This application addresses the situation where the total power supply is less than the total power demand by constructing a multi-objective optimization scheduling model to make decisions on energy consumption strategies for the next hour (i.e., a scheduling period T = 60 minutes, divided into several time segments, such as 4 time segments of 15 minutes each). The specific process is as follows: First, construct the objective function: ,in, This represents the total price of electricity purchased from the grid during time period t (the amount of electricity purchased multiplied by the average electricity price during time period t). The cost of purchasing gas from the gas network during period t (the amount of gas purchased multiplied by the average price of natural gas for electricity during period t). The demand response compensation cost is calculated by multiplying the sum of the dispatchable and replaceable electricity within time period t by a pre-set unit compensation price. This is used to incentivize users of transferable loads to adjust their energy consumption periods (pre-set by the park). Represents the function to be minimized. represents the objective function value, and s represents the number of time periods. If the power supply side of the park's energy system is greater than the total power demand side, then the corresponding electricity and gas purchases are 0. The constraint condition is that its electricity, heat, and gas demands are directly met by the energy supply at the current moment, that is, the corresponding total power supply side equals the total power demand side, which ensures that the total electricity consumption of each transferable load in the entire scheduling cycle is equal to the total electricity demand set by the user, and that it only operates within the time interval allowed by the user.
[0041] The energy allocation constraint for alternative loads requires the optimized dispatch system to determine the ratio of electricity and natural gas consumption in each time period t. Using the balance between total power supply and total power demand as a constraint, and by setting a gas-to-electricity equivalent coefficient, for each alternative load, the sum of its electricity consumption and equivalent gas consumption in each time period equals the original total energy demand forecast for that time period. Specifically, the total energy demand of the alternative load Q in time period t is set as follows: ,in, The amount of electricity consumed, The amount of gas consumed. This is the gas-to-electricity equivalence coefficient (i.e., the efficiency of converting unit natural gas calorific value into electrical energy). This is the product of the calorific value of natural gas and the thermoelectric conversion efficiency of the equipment, expressed in kWh / m³ (obtained from equipment ledgers). Constraints: , , Indicates the maximum electrical power of the load equipment. This indicates the maximum gas power (which can be converted into equivalent electrical power through calorific value).
[0042] For transferable load equipment (in this application, transferable load is referred to as "discrete interruptible load"), its total power consumption is fixed within the scheduling cycle, but the power consumption period can be shifted. Let the total power demand of the transferable load R within the scheduling cycle (the next hour) be W (collected through the park's energy management APP or reservation system, gathering user-defined equipment energy demands (such as electric vehicle charging, washing machine reservation periods), and summarizing the total power demand of various transferable loads within the scheduling cycle), then it must meet the following requirements: Where P is the rated power of the device, the user-defined available time interval of the device is obtained, and the start / stop status variables are limited. It takes the value 0 or 1 only within this time interval, and always takes the value 0 outside the time interval. , where 0 indicates the device is not started, and 1 indicates the device is started. The duration of the time period.
[0043] The collected data (photovoltaic forecasts, base load forecasts), along with generated load type labels, real-time electricity / gas prices, and equipment physical parameters, are used as inputs. A heuristic algorithm (particle swarm optimization in this embodiment) is employed to solve the problem. The outputs are the electricity purchase plans from the grid and the gas purchase plans from the gas grid for each time period within a preset scheduling cycle, electricity / gas allocation instructions for each alternative load, and start / stop time instructions for each transferable load. The model outputs are used for energy dispatch management within the park.
[0044] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description; sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.
[0045] 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 of the technical features. Such 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, and should all be included within the protection scope of this application.
Claims
1. A comprehensive energy optimization and scheduling system for industrial parks based on multi-energy flow coordination, characterized in that: The system includes: The integrated energy dispatch data preparation module of the park obtains the photovoltaic output benchmark value of the park based on the clear sky model, and compares it with the actual photovoltaic output of the park to obtain the output coefficient; it drives the LSTM neural network to predict the future output coefficient based on the meteorological data of the park, and combines it with the photovoltaic output benchmark value to obtain the final photovoltaic predicted output. The park energy load judgment module is used to analyze the electricity consumption periods of each load device in the park during the statistical period. Based on the overlap duration and number of overlap periods, the load time fluctuation of each load device is determined. Based on the theoretical maximum flow rate of the park's gas pipeline at the current pressure, combined with the predicted total gas load of the park in the future scheduling period, the adequacy of natural gas supply is determined. The similarity between the natural gas supply adequacy and the load power of the load devices in the historical statistical period is analyzed. Combined with the load time fluctuation, the substitution elasticity coefficient of the load devices is obtained, and the load type of the load devices is determined. The integrated energy dispatch module of the park is used to construct an optimal dispatch model with the goal of minimizing total cost. Based on the power balance of the supply and demand sides of the park, as well as the start-stop status and total power demand of transferable loads during the dispatch cycle, the constraints of the optimal dispatch model are obtained, and the optimal energy dispatch scheme is obtained by solving the problem. The determination of the load time fluctuation of each load device specifically involves: Obtain the daily electricity consumption period of each load device, overlap the daily electricity consumption periods of different dates within the set statistical period, and take the overlapping time intervals covered by the electricity consumption periods of each date as the characteristic time period of each load device. The load time fluctuation of each load device is obtained by summing the reciprocals of the number of electricity consumption periods corresponding to each characteristic period. The obtained replacement elasticity coefficient of the load equipment is specifically as follows: The correlation between load power and natural gas supply adequacy over a historical statistical period is obtained and denoted as p; Alternating elasticity coefficient The specific formula is: Where max() represents the maximum value function, and T represents the normalized result of the load time fluctuation of the load equipment.
2. The integrated energy optimization and scheduling system for industrial parks based on multi-energy flow collaboration as described in claim 1, characterized in that, The specific steps for obtaining the output coefficient include: When the photovoltaic power output benchmark value is less than the preset clear sky benchmark threshold, the power output coefficient k is set to 0; when the photovoltaic power output benchmark value is greater than or equal to the preset clear sky benchmark threshold, the ratio of the actual photovoltaic power output to the photovoltaic power output benchmark value is used as the power output coefficient.
3. The integrated energy optimization and scheduling system for industrial parks based on multi-energy flow collaboration as described in claim 1, characterized in that, The final photovoltaic power output is specifically the product of the photovoltaic power output benchmark value and the predicted future power output coefficient.
4. The integrated energy optimization and scheduling system for industrial parks based on multi-energy flow collaboration as described in claim 1, characterized in that, The process of determining the adequacy of natural gas supply is as follows: The theoretical maximum flow rate at the current inlet pressure of the gas pipeline network access point in the park is used as the upper limit of the gas supply capacity of the gas pipeline network in the corresponding statistical period; the total gas load forecast value is obtained by using a prediction model based on the historical gas consumption of the gas pipeline network in the park. The ratio of the upper limit of the gas supply capacity of the park's gas pipeline network to the predicted value of the total gas load is used as the sufficiency of natural gas supply.
5. The integrated energy optimization and scheduling system for industrial parks based on multi-energy flow collaboration as described in claim 1, characterized in that, The determination of the load type of the load equipment specifically includes: For each load device, if the normalized result of the load time fluctuation is greater than or equal to the first characteristic threshold, the load type of the load device is a transferable load. If the normalized result of the load time fluctuation is less than the first characteristic threshold, the correlation is less than 0, and the substitution elasticity coefficient is greater than or equal to the second characteristic threshold, the load type of the load equipment is a substitutable load; otherwise, it is a rigid load. Specifically, the first characteristic threshold is the average value of the load time fluctuation of all load equipment, and the second characteristic threshold is the average value of the substitution elasticity coefficient of all load equipment.
6. The integrated energy optimization and scheduling system for industrial parks based on multi-energy flow collaboration as described in claim 5, characterized in that, The objective function of the optimized scheduling model is specifically: ;in, This represents the total price of electricity purchased from the grid during time period t. The cost of purchasing gas from the gas network during time period t. To compensate for the cost of demand response, Represents the function to be minimized. Represents the objective function value. s This indicates the number of time periods; the time periods are determined by equally dividing a preset scheduling cycle.
7. The integrated energy optimization and scheduling system for industrial parks based on multi-energy flow collaboration as described in claim 6, characterized in that, The constraints for obtaining the optimized scheduling model are specifically as follows: With the balance between the total power supply and the total power demand as a constraint, by setting the gas-to-electricity equivalent coefficient, for each replaceable load, the sum of the electricity consumed in each time period and the equivalent gas volume is equal to the original total energy demand forecast value for that time period. Ensure that the total power consumption of each transferable load throughout the entire scheduling cycle is equal to the total power demand set by the user, and that it only operates within the time intervals allowed by the user.
8. The integrated energy optimization and scheduling system for industrial parks based on multi-energy flow collaboration as described in claim 7, characterized in that, The energy dispatch scheme includes: electricity purchase plans from the power grid for each time period within the preset dispatch cycle, gas purchase plans from the gas grid for each time period within the preset dispatch cycle, electricity / gas allocation instructions for each alternative load, and start / stop time instructions for each transferable load.