Energy storage method for district energy system based on flexible heat network technology

Abstract

The present application belongs to the field of comprehensive energy control, and particularly relates to a regional energy system energy storage method based on flexible heat network technology, which accurately predicts the power generation, energy demand and redundancy of each region at different time periods by comprehensively analyzing the historical energy storage device working state, energy demand and environmental data, combined with a seasonal prediction model, and then uses an energy configuration model to realize intelligent optimization of the regional energy storage and release amount; a distributed reinforcement control model based on a federal algorithm is constructed to ensure efficient storage and release of the energy storage device and energy scheduling; the performance of the energy storage device and the energy supply satisfaction rate are evaluated in real time, and the energy storage and release strategy is dynamically adjusted through a comprehensive evaluation score feedback mechanism; the present application improves the adaptability and response capability of the regional energy system to different environments through different types of energy storage modes.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of comprehensive energy control, and particularly relates to an energy storage method for regional energy systems based on flexible heat network technology. BACKGROUND

[0002] At present, with large-scale grid connection of renewable energy such as wind power, the intermittency and uncertainty of wind power output bring great challenges to the power system. At the same time, traditional thermal power equipment has limitations in peak shaving and flexibility, and it is difficult to meet the balance between increasing heating and power supply demands. In the prior art, although attempts have been made to combine wind power and thermal power operation, the flexibility of regional heat network energy storage and the diversity of user load characteristics are often ignored, resulting in that the overall benefit of the system cannot be maximized.

[0003] A patent with the authorized announcement number CN107910894B discloses a control method, device and system for combined operation of wind power equipment and thermal power equipment, comprising: determining the constraint conditions of the thermal power and electric power of the thermal power equipment, the constraint condition of the electric power of the wind power equipment, and the constraint condition of the sum of the electric power of the wind power equipment and the thermal power equipment based on the equipment parameters; determining the energy storage constraint condition of the regional heat network based on the energy storage parameters; under the premise of meeting the above constraint conditions, determining the electric power of the wind power equipment, the thermal power and electric power of the thermal power equipment, taking the thermal load demand of the thermal power equipment as the premise and taking the maximization of the system benefit of the combined operation of the wind power equipment and the thermal power equipment as the target; sending the wind power output consumption instruction to the wind power equipment and sending the dispatching instruction to the thermal power equipment.

[0004] A patent with the authorized announcement number CN113762643B discloses an energy storage capacity optimization configuration method for a regional comprehensive energy system, comprising: obtaining the grid operation parameters of a region to be analyzed; constructing a regional comprehensive energy system energy storage capacity optimization configuration objective function and constraint conditions; solving the objective function under the constraint conditions; and completing the energy storage capacity optimization configuration of the regional comprehensive energy system according to the solving result.

[0005] The above prior art has the following problems: the prior art only considers energy storage from the source end or considers comprehensive demand response or meets the requirements of electric and thermal energy coordinated dispatching and wind power consumption from the load end, but cannot accurately store and release different types of energy and schedule between regions according to the actual needs of the actual region. Therefore, the present application provides an energy storage method for regional energy systems based on flexible heat network technology. SUMMARY

[0006] In view of the deficiencies of the prior art, the present application provides an energy storage method for a regional energy system based on a flexible heat network technology, which accurately predicts the power generation, energy demand and surplus of each region at different time periods by comprehensively considering the historical working state of the energy storage device, energy demand and environmental data, in combination with a seasonal prediction model, and then uses an energy configuration model to realize intelligent optimization of the energy storage and release of the regional energy system; a distributed reinforcement control model based on a federal algorithm is used to ensure efficient storage and release of the energy storage device and energy scheduling; the performance of the energy storage device and the energy supply satisfaction rate are evaluated in real time, and the energy storage and release strategy is dynamically adjusted through a comprehensive evaluation score feedback mechanism; and the adaptability and response capability of the regional energy system to different environments are improved through different types of energy storage modes.

[0007] To achieve the above object, the present application provides the following technical scheme:

[0008] The energy storage method for a regional energy system based on a flexible heat network technology comprises the following steps:

[0009] S1, historical working state data of different regional energy storage devices, energy demand data and environmental data are obtained, and a prediction model with seasonal characteristics is configured to obtain the total power generation, energy demand and energy surplus of each region at different time periods;

[0010] S2, the total power generation, energy demand and energy surplus of each region at different time periods are used to obtain the energy storage and release of different types of energy of each region at corresponding time periods through an energy configuration model;

[0011] S3, a distributed reinforcement control model is constructed based on a federal algorithm, and the reinforcement control model is configured into the regional energy system, and the energy storage and release of different types of energy are controlled through the regional energy system to control the energy storage and release and scheduling of different types of energy of the energy storage device in different regions;

[0012] S4, a comprehensive score threshold is set, the storage and release and scheduling process parameters of the energy storage device of each region, the energy supply satisfaction rate data of the same region per unit time are obtained in real time, and a regional storage comprehensive evaluation score is obtained through an evaluation algorithm, when the regional storage comprehensive evaluation score is less than the comprehensive score threshold, the regional storage comprehensive evaluation score is fed back to the energy configuration model to optimize and adjust the energy storage and release and scheduling process of each region at corresponding time periods, otherwise no adjustment is made.

[0013] Specifically, the step of obtaining the energy surplus of each region at different time periods comprises:

[0014] S101, the electric-thermal energy consumption and temperature data of different regions with a time length of one year are obtained and preprocessed;

[0015] S102, input the preprocessed data into the clustering algorithm, cluster the regions, and obtain regions with the same energy demand data and temperature change trend;

[0016] S103, align the electrical energy consumption and thermal energy consumption corresponding to the same change trend region in the time dimension, and average the aligned electrical energy consumption and thermal energy consumption respectively to obtain the average electrical energy consumption and average thermal energy consumption time series corresponding to the same change trend region;

[0017] S104, according to the average electrical energy consumption and average thermal energy consumption time series, obtain the average electrical energy consumption change rate and average thermal energy consumption change rate sequence;

[0018] S105, input the average electrical energy consumption change rate and average thermal energy consumption change rate sequence into the clustering algorithm, and sequentially obtain K time period length intervals with the same electrical energy consumption and thermal energy consumption change rate and the corresponding change rate value.

[0019] Specifically, the step of obtaining the energy redundancy of each region in different time periods further comprises:

[0020] S106, using the time period length interval with the same change rate as the segmentation window, segmenting the corresponding region electrical energy consumption, thermal energy consumption and temperature sequence in the time dimension, and taking the corresponding change rate value as the seasonal factor to form the segmented input sequence pair (Ti, Xi, λi) and (Tj, Yj, γj) with the segmented electrical energy consumption and thermal energy consumption sequence, wherein Xi represents the i-th electrical energy consumption segmentation sequence, Ti represents the temperature segmentation sequence corresponding to the i-th electrical energy consumption segmentation sequence; λi represents the seasonal factor corresponding to the i-th electrical energy consumption segmentation sequence; Yj represents the j-th thermal energy consumption segmentation sequence, Tj represents the temperature segmentation sequence corresponding to the j-th thermal energy consumption segmentation sequence, and γj represents the seasonal factor corresponding to the j-th thermal energy consumption segmentation sequence;

[0021] S107, based on the integrated algorithm and the BiLSTM algorithm, an integrated prediction model is constructed, (Ti, Xi, λi) is input into BiLSTM1 in the prediction model, and (Tj, Yj, γj) is input into BiLSTM2 in the prediction model for training to obtain a trained prediction model;

[0022] S108, obtain the electrical-thermal energy consumption and temperature of each region in the previous 3 days of the current time period, and obtain the electrical-thermal demand of each region in the corresponding time period by the trained prediction model.

[0023] Specifically, the step of obtaining the energy redundancy of each region in different time periods comprises:

[0024] S109, obtaining historical total power generation and total heat supply data under different regional and environmental conditions, and fine-tuning the integrated prediction model pre-trained by using the obtained data, and simultaneously predicting the total power generation and total heat supply in the time period corresponding to S108;

[0025] S110, obtaining the energy shortage corresponding to power supply and heat supply in different regions in the corresponding time period based on the obtained total power generation and total heat supply and the predicted electric-thermal demand in the corresponding time period of each region; and represents the electric energy shortage corresponding to the length interval of the kth time period of the mth region, represents the thermal energy shortage corresponding to the length interval of the kth time period of the mth region.

[0026] Specifically, the steps of storage and dispatch of different types of energy include:

[0027] When the current time period , all regional energy storage devices enter the energy storage mode, and the specific steps of the energy storage process include:

[0028] S2011, obtaining the remaining capacity of the corresponding electric storage and heat storage devices of each region in the current time period, the cost of unit time electric storage and heat storage, and the unit time electric-thermal conversion cost Q P→H or the heat-thermal conversion cost Q H→P ;

[0029] S2012, when , if and , directly store and into the corresponding regional electric storage and heat storage devices, wherein and represent the remaining space of the electric storage device and the remaining space of the heat storage device of the mth region in the kth time period, respectively; represents the operating cost of the corresponding electric storage device of the mth region in the kth time period, represents the operating cost of the corresponding heat storage device of the mth region in the kth time period;

[0030] S2013, if and , convert the electric energy in the current time period into thermal energy and store it in the heat storage device until the heat storage device is full;

[0031] S2014, if there is surplus electric energy and there is an m! region satisfying , obtain the dispatch cost of dispatching thermal energy from the mth region to the m! region through the heat network model if and If yes, the remaining thermal energy is stored in the mth region through the enhanced control model, otherwise the remaining thermal energy is stored in the mth region corresponding to the power storage device; m! represents any region in the M regions except the mth region;

[0032] S2015, if the thermal-to-electricity conversion is repeated from S2012 to S2014, and the corresponding type of energy storage or dispatch storage is performed.

[0033] Specifically, the steps of storage and dispatch of different types of energy also include:

[0034] S2016, when if and the is stored in the mth region corresponding to the power storage device, and the thermal energy dispatching is determined through the heat network model, if and the remaining thermal energy is stored in the mth region through the enhanced control model, otherwise the heat is abandoned;

[0035] S2017, when and if the storage of the mth region corresponding to the power storage device is completed, if the thermal-to-electricity conversion is performed, and the converted electrical energy is stored in the mth region corresponding to the power storage device;

[0036] S2018, when the mth region corresponding to the power storage device cannot meet the corresponding electrical quantity of the thermal-to-electricity conversion process, the remaining converted electrical quantity is stored in the mth region corresponding to the power storage device, until all the remaining power storage devices meet , and the remaining heat is abandoned;

[0037] S2019, when if the thermal-to-electricity conversion storage is repeated by repeating S2017 and S2018, otherwise the thermal energy dispatch storage is repeated by repeating S2016;

[0038] S2020, when the is stored in the power storage device corresponding to the region, and the thermal energy dispatching is determined through S2016, if and the remaining thermal energy is stored in the mth region through the enhanced control model, and the excess thermal energy is abandoned, if not, when the mth region heat storage device state is , the remaining thermal energy is abandoned.

[0039] S2021, when and are satisfied, then repeat S2020 to perform thermal energy storage scheduling;

[0040] S2022, when , repeat the processes of S2016-S2021 to perform electric energy and thermal energy storage and scheduling judgment;

[0041] S2023, when , repeat the processes of S2016-S2022 to perform electric energy and thermal energy storage and scheduling judgment.

[0042] Specifically, the steps of storage-discharge and scheduling of different types of energy also include:

[0043] S2024, when the current time period or , then obtain the region, type and energy shortage amount of energy shortage through the judgment model;

[0044] S2025, if there is at least one region lacking electric energy in the current time period and all regions satisfy , then compare the electric energy shortage amount of the corresponding region with the remaining electric energy in the electric energy storage device of the corresponding region, if the remaining electric energy in the electric energy storage device is greater than the electric energy shortage amount, then compare the electric energy scheduling cost with the electric energy storage device operation cost, if the electric energy scheduling cost in the current time period is greater than the electric energy storage device operation cost and the thermal to electric conversion cost is greater than the electric energy storage device operation cost, then operate the electric energy storage device to discharge to supplement the missing electric energy;

[0045] S2026, if the electric energy scheduling cost in the current time period is greater than the electric energy storage device operation cost and the thermal to electric conversion cost is less than or equal to the electric energy storage device operation cost, then convert the corresponding thermal energy into electric energy to supplement the missing electric energy, if the converted electric energy is less than the missing electric energy, then operate the electric energy storage device to discharge to supplement;

[0046] S2027, when the electric energy scheduling cost in the current time period is less than the electric energy storage device operation cost and the thermal to electric conversion cost is greater than the electric energy storage device operation cost, if there is an mth region corresponding to , then perform electric energy scheduling supplement through the built-in distributed reinforcement control model, when the sum of the electric energy scheduled by all regions satisfying is less than the electric energy shortage amount, then operate the corresponding electric energy storage device to perform secondary supplement of the missing electric energy after scheduling supplement;

[0047] S2028. When the current time period's power dispatch cost is less than the energy storage device's operating cost and the heat-to-electricity cost is less than or equal to the energy storage device's operating cost, the power dispatch cost is compared with the heat-to-electricity cost. If the power dispatch cost is greater than the heat-to-electricity cost, the corresponding area's heat energy is converted into electricity to supplement the power shortage. If the supplement is insufficient to meet the corresponding power shortage, power dispatch is carried out through the S2027 process until the power shortage is met.

[0048] S2029. If the power dispatch cost is less than or equal to the heat-to-electricity conversion cost, then power dispatch is directly supplemented through process S2027. If the supplementation is insufficient to meet the corresponding power shortage, then excess heat energy in the same area is addressed through process S2026. If the thermal-to-electricity conversion is still insufficient to meet the power shortage, the sum of the costs of thermal-to-electricity conversion and power dispatch is compared with the operating cost of the energy storage device. If the sum is less, the missing power is supplemented by combining the thermal-to-electricity conversion and power dispatch processes. If the sum is greater, the energy storage device in the power-deficient area is operated to supplement the missing power after thermal-to-electricity conversion, until the power shortage is met.

[0049] Specifically, the steps for storing, storing, and dispatching different types of energy also include:

[0050] S2030. If the remaining energy in the energy storage device is less than or equal to the energy shortage, repeat the heat-to-electricity and energy dispatch process of S2026-S2029 to replenish the energy. If the energy replenished by heat-to-electricity and energy dispatch and the sum of the energy in the energy storage devices of all regions in the current time period are less than the energy shortage in the current time period, then feed back the replenished energy shortage to the regional energy system to issue a power shortage warning, and dispatch the power generation device through the regional energy system to carry out secondary power generation to replenish the energy.

[0051] S2031, If ​​at least one region in the current time period is missing thermal energy and all regions satisfy the following conditions: Then the S2024-S2030 process is repeated to replenish thermal energy.

[0052] S2032. If at least one storage area is missing thermal energy and electrical energy in the current time period, then repeat the process of S2024-S2031 to supplement the energy through dual scheduling of electrical energy and thermal energy.

[0053] S2033. If, after scheduling and replenishment during the S2024-S2032 process, the corresponding electrical or thermal energy in any region still meets the requirements... or If the conditions are met, the corresponding type of energy storage or disposal process in S2012-S2023 will be used to store or discard electrical or thermal energy in the corresponding area.

[0054] A computer readable storage medium, having stored thereon computer instructions which, when executed, perform a flexible thermal network technology-based regional energy system energy storage method.

[0055] Compared with the prior art, the beneficial effects of the present application are:

[0056] The present application aims at the deficiencies of the prior art, and through the application of historical data and the prediction model with seasonal characteristics, the present application can accurately estimate the energy supply and demand of each region in a specific time period, providing solid data support for optimal allocation. Secondly, through fine calculation of the energy storage and release amount of each region, efficient management of energy storage and release is realized. In addition, with the help of a distributed reinforcement control model, the present application enhances the precise control and scheduling of energy storage devices, improves the response speed and coordination of the system. By setting a comprehensive score threshold and real-time evaluating the operation effect of the energy storage device, the present application can quickly feedback the optimization demand, ensuring that the energy allocation is in the best state. BRIEF DESCRIPTION OF DRAWINGS

[0057] Figure 1 The present application provides a flexible thermal network technology-based regional energy system energy storage method flowchart for embodiment 1;

[0058] Figure 2 The present application provides a prediction model structure flowchart for embodiment 1. DETAILED DESCRIPTION

[0059] Embodiment 1

[0060] Please refer to Figure 1 The present application provides a flexible thermal network technology-based regional energy system energy storage method, which comprises the following steps:

[0061] S1, obtaining historical data of different regional energy storage device working state, energy demand data and environmental data, and through a configured prediction model with seasonal characteristics, obtaining the total power generation, energy demand and energy shortage of each region in different time periods;

[0062] Further, please refer to Figure 2 The present application provides a flexible thermal network technology-based regional energy system energy storage method, which comprises the following steps:

[0063] S101, obtaining and preprocessing the electric-thermal energy consumption and temperature data of different regions with a time length of one year;

[0064] S102, inputting the preprocessed data into a clustering algorithm to cluster the regions, and obtaining regions with the same energy demand data and temperature change trend;

[0065] S103, align the electricity consumption and heat consumption corresponding to the same change trend area in the time dimension, and average the aligned electricity consumption and heat consumption respectively to obtain average electricity consumption and average heat consumption time series corresponding to the same change trend area;

[0066] S104, obtain average electricity consumption change rate and average heat consumption change rate sequence according to the average electricity consumption and average heat consumption time series;

[0067] S105, input the average electricity consumption change rate and the average heat consumption change rate sequence into a clustering algorithm to obtain each time period length interval with the same electricity consumption and heat consumption change rate and the corresponding change rate value in turn.

[0068] S106, use the time period length interval with the same change rate as a segmentation window to segment the corresponding area electricity consumption, heat consumption and temperature sequence in the time dimension, and use the corresponding change rate value as a seasonal factor to form the segmented input sequence pair (Ti, Xi, λi) and (Tj, Yj, γj) with the segmented electricity consumption and heat consumption sequence, wherein Xi represents the i-th electricity consumption segmented sequence, Ti represents the temperature segmented sequence corresponding to the i-th electricity consumption segmented sequence; λi represents the seasonal factor corresponding to the i-th electricity consumption segmented sequence; Yj represents the j-th heat consumption segmented sequence, Tj represents the temperature segmented sequence corresponding to the j-th heat consumption segmented sequence, and γj represents the seasonal factor corresponding to the j-th heat consumption segmented sequence;

[0069] S107, construct an integrated prediction model based on an integrated algorithm and a BiLSTM algorithm, input (Ti, Xi, λi) into BiLSTM1 in the prediction model, and input (Tj, Yj, γj) into BiLSTM2 in the prediction model for training to obtain a trained prediction model;

[0070] S108, obtain the electricity-heat consumption and temperature of each area in the previous 3 days of the current time period, and predict the electricity-heat demand in the corresponding time period of each area through the trained prediction model.

[0071] S109, obtain historical total power generation and total heat supply data under different environmental conditions of different areas, and fine-tune the pre-trained integrated prediction model using the obtained data, and simultaneously predict the total power generation and total heat supply in the time period corresponding to S108;

[0072] S110, based on the obtained total power generation and total heat supply and the predicted electricity-heat demand in the corresponding time period of each area, obtain the energy redundancy corresponding to power supply and heat supply in turn in the corresponding time of different areas and represents the energy surplus of the mth region in the kth time period length interval, represents the energy surplus of the mth region in the kth time period length interval.

[0073] The process identifies regions with similar energy consumption patterns through preprocessing and clustering analysis of historical data, improving data comparability and model accuracy. Subsequently, by calculating the average consumption change rate and seasonal factor, the dynamic characteristics of energy demand are refined; BiLSTM is used for deep learning training, enabling the model to handle sequence data and enhance the fitting degree of nonlinear time series, thereby improving prediction accuracy; by fine-tuning the pre-trained model, the model parameters are further optimized to better adapt to actual environmental changes; finally, based on the model prediction results, the energy surplus of each region can be accurately obtained, guiding energy scheduling and avoiding excessive power generation or insufficient heating, achieving fine management and efficient use of energy.

[0074] S2, according to the total power generation, energy demand and energy surplus of each region in different time periods, through the energy allocation model, the different types of energy storage and release amount of each region corresponding to the time period are obtained;

[0075] Further, the energy allocation model in the embodiment is constructed by a random forest algorithm.

[0076] S3, based on the federated algorithm, a distributed reinforcement control model is constructed, and the reinforcement control model is configured to the regional energy system, according to the different types of energy storage and release amount, the energy storage and release and scheduling of different types of energy in different regions are controlled through the regional energy system;

[0077] Further, the reinforcement control model in the embodiment is constructed by SAC algorithm;

[0078] Further, the steps of energy storage and release and scheduling of different types of energy in the embodiment include:

[0079] When the current time period , all regional energy storage devices enter the energy storage mode, and the energy storage process includes the following steps:

[0080] S2011, the remaining capacity of the corresponding energy storage and release device of each region in the current time period, the cost of energy storage and release per unit time, and the cost of electricity to heat Q P→H or heat to electricity Q H→P per unit time are obtained.

[0081] S2012, when the following condition is met , if and , then directly and stores into the corresponding area electricity storage and heat storage device, wherein and indicates the remaining space of the electricity storage device and the remaining space of the heat storage device corresponding to the mth area in the kth time period in turn; indicates the operating cost of the electricity storage device corresponding to the mth area in the kth time period, indicates the operating cost of the heat storage device corresponding to the mth area in the kth time period;

[0082] S2013, if and , the current time period electric energy is converted into heat energy, and is stored in the heat storage device until the heat storage device is full;

[0083] S2014, if the power is surplus and there is an mth area satisfying , the scheduling cost of dispatching heat energy from the mth area to the mth area is obtained through the heat network model if and , the remaining heat energy is dispatched to the mth area for storage through the reinforcement control model, otherwise the remaining power is stored in the electricity storage device corresponding to the mth area; mth area in M areas;

[0084] S2015, if , repeat S2012 to S2014 to judge heat-to-electricity, and store and dispatch the corresponding type of energy. Further, the dispatching storage in S2014 is to dispatch the remaining heat energy from the mth area to the mth area for storage through the heat network model and the reinforcement control model.

[0085] S2016, when , if , store in the electricity storage device corresponding to the mth area and judge the heat energy dispatching through the heat network model, if and , the remaining heat energy is dispatched to the mth area for storage through the reinforcement control model, otherwise heat is abandoned;

[0086] S2017, when and , if is stored, the electricity storage device corresponding to the mth area satisfies , heat-to-electricity operation is performed, and the converted electric energy is stored in the electricity storage device corresponding to the mth area;

[0087] S2018, when the mth region corresponding to the power storage device cannot meet the corresponding power of the heat-to-power conversion process, the remaining converted power is stored in the mth region corresponding to the power storage device, until all the remaining power storage devices meet , and the remaining heat is discarded for heat treatment;

[0088] S2019, when and , if , repeat the S2017 and S2018 processes to store heat-to-power, otherwise repeat the S2016 process to store heat energy;

[0089] S2020, when and , store in the corresponding region of the power storage device, and perform heat energy scheduling through S2016, if and , schedule the remaining heat energy to the mth region for storage through the reinforcement control model, and discard the excess heat energy for heat treatment, if not, when the mth region heat storage device state is , discard the remaining heat energy for heat treatment;

[0090] S2021, when and , repeat S2020 to store heat energy;

[0091] S2022, when , repeat the S2016-S2021 process to store and schedule electrical and thermal energy;

[0092] S2023, when , repeat the S2016-S2022 process to store and schedule electrical and thermal energy.

[0093] S2024, when the current time period or , obtain the region, type and energy shortage amount of the energy shortage through the discrimination model;

[0094] S2025, if at least one region lacks electrical energy in the current time period and all regions meet , compare the electrical energy shortage amount of the corresponding region with the remaining electrical energy in the corresponding power storage device, if the remaining electrical energy in the power storage device is greater than the electrical energy shortage amount, compare the electrical energy scheduling cost with the power storage device operation cost, if the electrical energy scheduling cost is greater than the power storage device operation cost and the heat-to-power cost is greater than the power storage device operation cost in the current time period, operate the power storage device to discharge to supplement the missing electrical energy;

[0095] S2026, if the current time period electric energy scheduling cost is greater than the storage device operating cost and the heat-to-electricity cost is less than or equal to the storage device operating cost, the corresponding regional heat energy is converted into electric energy, and the electric energy is supplemented, if the corresponding electric energy after conversion is less than the missing electric energy, the storage device is operated to discharge and supplement;

[0096] S2027, when the current time period electric energy scheduling cost is less than the storage device operating cost and the heat-to-electricity cost is greater than the storage device operating cost, if there is an mth region corresponding to the electric energy scheduling supplement is carried out through the built-in distributed reinforcement control model, when the sum of the electric energy scheduling of all regions satisfying is less than the electric energy missing amount, the corresponding storage device is operated to carry out secondary supplement of the missing electric energy after the scheduling supplement;

[0097] S2028, when the current time period electric energy scheduling cost is less than the storage device operating cost and the heat-to-electricity cost is less than or equal to the storage device operating cost, the electric energy scheduling cost and the heat-to-electricity cost are compared, if the electric energy scheduling cost is greater than the heat-to-electricity cost, the corresponding regional heat energy is converted into electric energy, and the electric energy is supplemented, if the supplement cannot meet the corresponding electric energy missing amount, the electric energy scheduling supplement is carried out through the S2027 process until the missing electric energy is met;

[0098] S2029, if the electric energy scheduling cost is less than or equal to the heat-to-electricity cost, the electric energy scheduling supplement is directly carried out through the S2027 process, if the supplement cannot meet the corresponding electric energy missing amount, the S2026 process is carried out to supplement the redundant heat energy of the same region through heat-to-electricity, if the electric energy missing amount cannot still be met, the sum of the heat-to-electricity and the electric energy scheduling cost is compared with the storage device operating cost, if it is less, the missing electric energy is supplemented through the joint heat-to-electricity and electric energy scheduling process, if it is greater, the storage device of the power shortage region is operated to supplement the missing electric energy after the heat-to-electricity supplement, until the missing electric energy is met.

[0099] S2030, if the remaining electric energy in the storage device is less than or equal to the electric energy missing amount, the heat-to-electricity and electric energy scheduling processes of the S2026-S2029 processes are repeated to supplement the electric energy, if the sum of the electric energy supplemented by the heat-to-electricity and electric energy scheduling and the sum of the electric energy in the storage device of all regions in the current time period is less than the electric energy missing amount in the current time period, the supplemented electric energy missing amount is fed back to the regional energy system to carry out power shortage warning, and the power generation device is dispatched through the regional energy system to carry out secondary power generation supplement;

[0100] S2031, if there is at least one region missing heat energy in the current time period and all regions satisfy the heat energy scheduling supplement is repeated through the S2024-S2030 process.

[0101] S2032, if the current time period stores at least one missing regional heat energy and electric energy, repeating the processes of S2024-S2031 to perform dual scheduling supplement of electric energy and heat energy;

[0102] S2033, if any regional electric energy or heat energy still satisfies the condition after the scheduling supplement in the processes of S2024-S2032, using the corresponding type energy storage or discard process in S2012-S2023 to perform storage or discard operation on the electric energy or heat energy of the corresponding region;

[0103] In the above processes, in the energy storage mode, the system ensures that the energy is stored in the most cost-effective way by monitoring the remaining capacity of the energy storage device and the related cost in real time, avoiding unnecessary energy waste. When there is excess electric energy, the system will decide whether to convert the electric energy into heat energy for storage or directly store the electric energy based on the state of the heat storage device. For the case of excess heat energy, the system will also select the most economical way to schedule and store heat energy based on cost-benefit analysis; in addition, in the energy scheduling stage, a distributed reinforcement learning control model is used, and the system can flexibly respond to changes in energy demand between different regions, select the most cost-effective energy supplement method by comparing the cost-benefit of electric energy scheduling and heat-to-electricity conversion, and ensure efficient use of energy; when there is energy shortage, the system can quickly identify the region and type of energy that needs to be supplemented, and select the most appropriate energy conversion or scheduling method to fill the gap by comparing the costs of different supplement schemes; for example, when the electric energy scheduling cost is higher than the energy storage device operation cost, the system will choose to run the energy storage device to discharge to supplement the missing electric energy; otherwise, it will supplement through electric energy scheduling; in the case of heat energy shortage, the system will also adopt similar strategies to ensure timely supplement of heat energy; in summary, through the above mechanisms, not only the energy utilization efficiency is significantly improved, the energy waste is reduced, but also the self-adaptive adjustment ability and the energy storage and deployment of different types of energy of the system are enhanced, which can adapt to more environmental states;

[0104] S4, setting a comprehensive score threshold, obtaining the storage-discharge and scheduling process parameters of each regional energy storage device, and the energy supply satisfaction rate data of the same region per unit time in real time, and obtaining the regional storage comprehensive evaluation score through the evaluation algorithm, when the regional storage comprehensive evaluation score is less than the comprehensive score threshold, the regional storage comprehensive evaluation score is fed back to the energy configuration model to optimize and adjust the energy storage-discharge amount and scheduling process of each region at the corresponding time period, otherwise no adjustment is made. Further, the evaluation algorithm in the embodiment is a comprehensive fuzzy evaluation algorithm.

[0105] Embodiment 2

[0106] ​​A computer readable storage medium, having stored thereon computer instructions which, when executed, perform a method for energy storage of a district energy system based on a flexible thermal grid technology.

[0107] An electronic device, comprising a memory and a processor, the memory storing a computer program, the processor implementing a method for energy storage of a district energy system based on a flexible thermal grid technology when executing the computer program.

[0108] The embodiments of the present application are described above with reference to the drawings, but the present application is not limited to the specific embodiments described above, and the specific embodiments described above are merely illustrative, but not restrictive, and a person of ordinary skill in the art can make changes, modifications, replacements and variations to the above embodiments without departing from the purpose of the present application and the scope protected by the claims, and these are all within the protection of the present application.

Claims

1. A method for energy storage in a district energy system based on flexible thermal grid technology, characterized in that, The method comprises the following steps: S1, obtaining historical different regional energy storage device working state data, energy demand data and environmental data, and obtaining total power generation, energy demand and energy surplus of each region in different time periods through a configured prediction model with seasonal characteristics; S2, obtaining different types of energy storage-discharge amounts of each region in the corresponding time period through an energy configuration model according to the total power generation, energy demand and energy surplus of each region in different time periods; S3, constructing a distributed reinforcement control model based on a federal algorithm, configuring the reinforcement control model to a regional energy system, and controlling the energy storage and discharge of different types of energy in different regions through the regional energy system according to the different types of energy storage-discharge amounts; S4, setting a comprehensive score threshold, obtaining the storage-discharge and scheduling process parameters of the energy storage device of each region, the energy supply satisfaction rate data of the same region per unit time, and obtaining the regional storage comprehensive evaluation score through an evaluation algorithm, when the regional storage comprehensive evaluation score is less than the comprehensive score threshold, the regional storage comprehensive evaluation score is fed back to the energy configuration model to optimize and adjust the energy storage-discharge amount and scheduling process of each region in the corresponding time period, otherwise no adjustment is made; The steps of the storage and scheduling of different types of energy comprise: When the current time period All regional energy storage devices enter energy storage mode, where M is the total number of regions, and the energy storage process includes the following steps: S2011. Obtain the remaining capacity of the energy storage and thermal storage devices for each region in the current time period, the cost of energy storage and thermal storage per unit time, and the cost of electricity-to-heat conversion per unit time. Or the cost of heat to electricity ; S2012、When S2011 is satisfied , if and , then directly store and into the corresponding regional storage and heat storage devices, wherein and represent the remaining space of the electricity storage device and the remaining space of the heat storage device corresponding to the length of the mth regional kth time period, respectively; represents the operation cost of the electricity storage device corresponding to the mth region in the kth time period, represents the operation cost of the heat storage device corresponding to the mth region in the kth time period; represents the electricity surplus corresponding to the length interval of the mth regional kth time period, represents the heat surplus corresponding to the length interval of the mth regional kth time period; S2013、if and the current time period is converted into heat energy and stored in the heat storage device until the heat storage device is full. S2014、if there is surplus power and there is a first region that satisfies , then through the heat network model, the scheduling cost of scheduling heat energy from the mth region to the first region is obtained , if and , then the surplus heat energy is scheduled to the first region for storage through the reinforcement control model, otherwise the surplus power is stored in the power storage device corresponding to the mth region; any region in the M regions except the mth region is represented S2015、if S2012 to S2014 for thermal conversion electricity judgment, and corresponding type energy storage or dispatch storage is carried out; S2016、When , if and are met, the stored in the energy storage device corresponding to the mth region is subjected to thermal energy scheduling judgment through the heat network model, if and are met, the remaining thermal energy is scheduled to the mth region for storage through the reinforcement control model, otherwise, heat is abandoned. ​ S2017, when satisfied and At that time, if After storage is completed, the energy storage device corresponding to the m-th region satisfies If the thermal-to-electrical conversion is performed, the converted electrical energy will be stored in the energy storage device corresponding to the m-th region. S2018、When the power storage device corresponding to the mth region cannot meet the corresponding power of the heat-to-power conversion process, the remaining converted power is stored in the power storage device corresponding to the mth region, until all the remaining power storage devices meet , and the remaining heat is discarded. ​ S2019, when and S2018, if , repeat the process of S2017 and S2018 for thermal conversion electrical storage, otherwise repeat the process of S2016 for thermal energy dispatch storage; S2020, when satisfied and Then The energy is stored in the corresponding energy storage device, and thermal energy dispatch is determined through S2016. If the conditions are met... and Then, by strengthening the control model, the remaining thermal energy will be dispatched to the first... The region stores the heat and discards excess heat. If this is not satisfied, the heat storage device in the m-th region is in a certain state. When this happens, the remaining heat energy will be discarded. S2021、when and S2020 is repeated to perform thermal energy scheduling storage. S2022、When S2016-S2021 are repeated to make a judgment on the storage and dispatch of electric energy and thermal energy. S2023、when S2016-S2022 are repeated to make a judgment on the storage and dispatch of electric energy and thermal energy. S2024、when the current time period or is, then the region, type and energy shortage amount of energy shortage are obtained by the discrimination model; S2025、if the current time period has at least one region missing electric energy and all regions satisfy comparing the corresponding region electric energy missing amount with the remaining electric energy in the corresponding region electric energy storage device, if the remaining electric energy in the electric energy storage device is greater than the electric energy missing amount, comparing the electric energy scheduling cost with the electric energy storage device operation cost, if the current time period electric energy scheduling cost is greater than the electric energy storage device operation cost and the thermal to electric conversion cost is greater than the electric energy storage device operation cost, operating the electric energy storage device to discharge and supplement the missing electric energy. S2026、if the current time period electric energy scheduling cost is greater than the storage device operation cost and the thermal to electric conversion cost is less than or equal to the storage device operation cost, the corresponding regional thermal energy is converted into electric energy, and the electric energy deficiency is supplemented, and if the converted corresponding electric energy is less than the electric energy deficiency, the storage device is operated to discharge and supplement. S2027、When the current time period energy scheduling cost is less than the energy storage device operation cost and the heat-to-power cost is greater than the energy storage device operation cost, if there is a region corresponding to the first , the built-in distributed reinforcement control model is used for energy scheduling supplement. When the sum of the energy of all regions that meet the scheduling is less than the energy deficiency, the corresponding energy storage device is operated to supplement the deficient energy after scheduling supplement. ; S2028, when the current time period electric energy scheduling cost is less than the storage device operation cost and the heat to electricity cost is less than or equal to the storage device operation cost, the electric energy scheduling cost and the heat to electricity cost are compared, if the electric energy scheduling cost is greater than the heat to electricity cost, the heat energy of the corresponding region is converted into electric energy to supplement the electric energy deficiency, if the supplemented electric energy cannot meet the corresponding electric energy deficiency, the electric energy scheduling supplement is carried out through the process of S2027 until the electric energy deficiency is met; S2029, if the electricity scheduling cost is less than or equal to the heat-to-power cost, directly perform electricity scheduling supplement through the S2027 process, if the supplement cannot meet the corresponding electricity deficiency, perform heat-to-power supplement through the S2026 process for the same region redundant heat energy , if the electricity deficiency still cannot be met, compare the sum of the heat-to-power and electricity scheduling costs with the storage device operation cost, if the sum is less, supplement the electricity deficiency through the joint heat-to-power and electricity scheduling process, if the sum is greater, operate the electricity deficiency region storage device, supplement the electricity deficiency after the heat-to-power supplement, until the electricity deficiency is met; S2030, if the remaining electric energy in the storage device is less than or equal to the electric energy deficiency, the heat to electricity and electric energy scheduling processes of S2026-S2029 are repeated to supplement the electric energy, if the supplemented electric energy and the sum of the electric energy in the storage device of all regions in the current time period are less than the electric energy deficiency in the current time period, the supplemented electric energy deficiency is fed back to the regional energy system to give an electric power shortage warning, and the power generation device is dispatched to carry out secondary power generation supplement through the regional energy system; S2031、if the current time period stores at least one missing heat energy of the region and all regions satisfy S2031, if the current time period stores at least one missing heat energy of the region and all regions satisfy S2031, if the current time period stores at least one missing heat energy of the region and all regions satisfy S2032, if at least one region in the current time period lacks heat energy and electric energy, the processes of S2024-S2031 are repeated to carry out double scheduling supplement of electric energy and heat energy; S2033, if the power or thermal energy of any region still satisfies or the condition after the dispatching and supplementing in S2024-S2032, then store or discard the power or thermal energy of the corresponding region using the corresponding type of energy storage or discarding process in S2012-S2023.

2. The district energy system energy storage method based on flexible thermal grid technology of claim 1, wherein, The steps of obtaining the energy surplus of each region in different time periods comprise: S101, obtaining electric-thermal energy consumption and temperature data of different regions with a time length of one year and preprocessing; S102, inputting the preprocessed data into a clustering algorithm to cluster the regions and obtain regions with the same energy demand data and temperature change trend; S103, aligning the electric energy consumption and heat energy consumption of the regions with the same change trend in the time dimension, and averaging the aligned electric energy consumption and heat energy consumption respectively to obtain the average electric energy consumption and average heat energy consumption time series of the regions with the same change trend; S104, obtaining a sequence of average electric energy consumption change rates and average thermal energy consumption change rates according to the sequence of average electric energy consumption and average thermal energy consumption; S105, inputting the sequence of average electric energy consumption change rates and average thermal energy consumption change rates into a clustering algorithm to sequentially obtain K time period length intervals with the same electric energy consumption and thermal energy consumption change rates and corresponding change rate values.

3. The district energy system energy storage method based on flexible thermal grid technology of claim 2, wherein, The step of obtaining the energy redundancy of each region in different time periods further comprises: S106, using the time period length interval with the same change rate as the segmentation window to segment the corresponding regional electricity consumption, heat consumption and temperature sequence in the time dimension, and taking the corresponding change rate value as the season factor to form the segmented input sequence pair with the segmented electricity consumption and heat consumption sequence and wherein represents the i-th electricity consumption segmentation sequence, represents the temperature segmentation sequence corresponding to the i-th electricity consumption segmentation sequence; represents the season factor corresponding to the i-th electricity consumption segmentation sequence; wherein represents the j-th heat consumption segmentation sequence, represents the temperature segmentation sequence corresponding to the j-th heat consumption segmentation sequence, represents the season factor corresponding to the j-th heat consumption segmentation sequence; S107、based on the integrated algorithm and the BiLSTM algorithm, an integrated prediction model is constructed, and input into BiLSTM1 in the prediction model, and simultaneously input into BiLSTM2 in the prediction model for training, and a trained prediction model is obtained; S108, obtaining the electric-thermal energy consumption and temperature of each region in the previous 3 days of the current time period, and obtaining the electric-thermal demand of each region in the corresponding time period by using the trained prediction model.

4. The district energy system energy storage method based on flexible thermal grid technology of claim 3, wherein, The step of obtaining the energy redundancy of each region in different time periods comprises: S109, obtaining historical total power generation and total heat supply data under different environmental conditions of different regions, and fine-tuning the pre-trained integrated prediction model by using the obtained data, and simultaneously predicting the total power generation and total heat supply in the time period corresponding to S108; S110, based on the obtained total power generation and total heat supply and the predicted electric-thermal demand of each region in the corresponding time period, obtaining the energy shortage corresponding to the power supply and heat supply of different regions in the corresponding time in turn and .

5. A computer readable storage medium, characterized in that, The computer program product has computer instructions stored thereon, and when the computer instructions are executed, the energy storage method of the district energy system based on the flexible heat network technology is executed.

Images