Aquifer heat storage peak regulation method and system using wind power abandoned electricity

Abstract

The invention relates to the technical field of aquifer heat storage and peak regulation, in particular to an aquifer heat storage and peak regulation method and system utilizing wind power abandoned electricity. The method comprises the following steps: calculating a hourly load prediction value based on acquired wind power abandoned power data, and extracting real-time abandoned power; based on the real-time electricity abandoning quantity and a preset electricity abandoning threshold value, a matching decision of electricity abandoning, heat storage and load supply is made; based on the obtained aquifer state data and hourly load prediction value, heat storage control and heat release control are carried out in combination with the result of the matching decision; according to the peak regulation instruction and the heat storage control and heat release control result, self-adaptive peak regulation control is carried out, and the peak regulation control result is calculated; efficient conversion and storage of abandoned wind power are achieved, the abandoned power consumption rate and the aquifer heat storage efficiency are improved, the scale limitation of electrochemical energy storage is broken through, the problem of abandoned power waste is solved, and stable heat source support is provided for power grid peak regulation in a low-cost and large-capacity heat storage mode.

Description

Technical Field

[0001] This application relates to the field of robot control technology, specifically to a method and system for peak shaving using aquifer thermal storage based on wind power curtailment. Background Technology

[0002] As the global energy structure accelerates its transformation towards cleaner and lower-carbon energy, wind power, as an important form of renewable energy, has seen its installed capacity and power generation share grow rapidly. However, the inherent intermittency and strong volatility of wind energy resources often make it difficult to match wind power output with the grid load curve. During periods when the grid's absorption capacity is insufficient, transmission channels are limited, or system peak-shaving resources are scarce, "wind curtailment" measures must be implemented to ensure the safe and stable operation of the grid, resulting in a significant waste of clean energy. At the same time, socio-economic development and changes in electricity consumption patterns have led to a widening gap between peak and valley loads on the grid, making the demand for fast, flexible, and economical peak-shaving resources increasingly urgent.

[0003] Currently, power grid peak shaving mainly relies on traditional thermal power unit regulation, gas turbine start-up and shutdown, and emerging electrochemical energy storage. Traditional thermal power peak shaving has a slow response speed, low efficiency and increased emissions during low-load operation, which contradicts the goals of energy conservation and emission reduction. While gas turbine peak shaving is relatively flexible, its operating fuel costs are high and it is significantly affected by fluctuations in gas supply and prices. Electrochemical energy storage technology, represented by lithium-ion batteries, has a rapid response, but its cost remains high, its cycle life is limited, and it is constrained by energy density, making it more suitable for short-term, high-frequency power regulation and unable to meet large-scale, long-term peak shaving needs. Therefore, exploring an energy storage peak shaving technology that combines large-scale, long-cycle, and low-cost characteristics has become a key breakthrough urgently needed by the industry.

[0004] Aquifer thermal energy storage technology, as a large-scale seasonal energy storage method, utilizes the vast space and heat capacity of underground aquifers to inject and store thermal energy, which can then be extracted and used when needed. It boasts significant advantages such as large storage capacity, long service life, and low unit cost, theoretically providing a highly promising solution for absorbing fluctuating renewable energy. However, existing technologies for using aquifers for thermal energy storage are mostly concentrated in solar thermal storage or industrial waste heat recovery, exhibiting significant disconnect and inadequacy in synergistic coupling with wind power curtailment, failing to resolve the core contradiction between wind power absorption and grid peak shaving.

[0005] The information disclosed in this background section is intended only to enhance the understanding of the overall background of this application and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this application is to provide a method for peak shaving by utilizing aquifer thermal storage using wind power curtailment, which to some extent solves the problems raised in the background technology, realizes efficient conversion and storage of wind power curtailment, and improves the curtailment absorption rate and aquifer thermal storage efficiency.

[0007] To achieve the above objectives, this application provides the following technical solution:

[0008] In a first aspect, this application provides a method for peak shaving using aquifer thermal storage by utilizing wind power curtailment, comprising the following steps:

[0009] S1: Real-time synchronous acquisition of wind power curtailment data and aquifer status data, and preprocessing of the wind power curtailment data and aquifer status data;

[0010] S2: Based on the wind power curtailment data, an hourly load forecast value is calculated using an LSTM model incorporating an attention mechanism, and the LSTM model is updated in real time;

[0011] S3: Extract the real-time abandoned power and set the abandoned power threshold; based on the real-time abandoned power and the abandoned power threshold, make matching decisions on abandoned power, thermal storage and load supply;

[0012] S4: Based on the aquifer state data and hourly load prediction values, and combined with the results of matching decisions, perform thermal storage control and heat release control;

[0013] S5: Receive the power grid peak shaving command; perform adaptive peak shaving control based on the peak shaving command and the results of the thermal storage control and heat release control, and calculate the result of the peak shaving control;

[0014] S6: Based on the results of the peak shaving control, repeat steps S3 to S5; when the preset iteration conditions are met, stop the iteration and output the peak shaving control results obtained in the last iteration.

[0015] As a preferred embodiment of the aquifer thermal storage peak-shaving method utilizing wind power curtailment described in this application, wherein: the wind power curtailment data includes real-time curtailment volume; and the thermal storage control steps specifically include:

[0016] Obtain the underground ambient temperature and calculate the temperature difference between the underground ambient temperature and the temperature of the thermal storage area, i.e., the underground ambient temperature minus the temperature of the thermal storage area;

[0017] The thermal conductivity of the aquifer is calculated by measuring the aquifer core sample in the laboratory, and the thermal conductivity of the aquifer is multiplied by the dimensionless temperature difference to obtain the first product;

[0018] Calculate the effective thickness of the thermal storage zone and obtain the porosity; multiply the dimensionless effective thickness of the thermal storage zone and the porosity by the second product;

[0019] Calculate the ratio of the first product to the second product to obtain the heat loss rate;

[0020] The electrothermal conversion efficiency is determined based on the type of electrothermal device. The real-time abandoned power is multiplied by the electrothermal conversion efficiency to obtain the thermal storage power.

[0021] As a preferred embodiment of the aquifer thermal storage peak-shaving method for utilizing wind power curtailment described in this application, the thermal storage control step further includes:

[0022] Obtain the remaining thermal storage capacity, specific heat capacity of the thermal storage medium, and allowable temperature rise of the thermal storage medium in the thermal storage zone; the thermal storage power is less than or equal to the cumulative product of the remaining thermal storage capacity, specific heat capacity of the thermal storage medium, and allowable temperature rise of the thermal storage medium.

[0023] Calculate the temperature fluctuation value of the thermal storage area, and set a compensation coefficient based on the temperature fluctuation value; the compensation coefficient is positively correlated with the temperature fluctuation value.

[0024] Calculate the cumulative product of the heat loss rate, thermal storage power, and compensation coefficient to obtain the thermal compensation power; obtain the initial thermal storage power, and add the thermal compensation power to the initial thermal storage power to obtain the actual thermal storage power;

[0025] Adjust the output current or voltage of the electrothermal conversion device to control the heat storage zone to store heat with the actual heat storage power.

[0026] As a preferred embodiment of the aquifer thermal storage peak-shaving method for utilizing wind power curtailment described in this application, the heat release control step specifically includes:

[0027] The current heat storage capacity is calculated based on the temperature, volume, and specific heat capacity of the heat storage medium in the heat storage area, and the remaining heat storage capacity is calculated based on the current heat storage capacity.

[0028] The remaining heat storage capacity is the difference between the remaining heat storage capacity and the current heat storage capacity, that is, the remaining heat storage capacity minus the current heat storage capacity.

[0029] Obtain the conventional heating power of the power grid, and calculate the load gap based on the conventional heating power of the power grid and the hourly load forecast value;

[0030] The load gap is the difference between the hourly load forecast and the grid's conventional heating power, i.e., the hourly load forecast minus the grid's conventional heating power.

[0031] As a preferred embodiment of the aquifer thermal storage peak-shaving method utilizing wind power curtailment described in this application, the heat release control step further includes:

[0032] If the heat storage margin is greater than or equal to the load gap, the heat storage margin is converted into actual heat release power through the heat release device of the aquifer.

[0033] The actual heat release power is dynamically adjusted according to the load gap.

[0034] If the thermal storage capacity is less than the load gap, an auxiliary heating request is sent to the power grid dispatch center to call on the heating reserve capacity of conventional power sources such as thermal power units and gas turbines to fill the remaining load gap.

[0035] Record load gap data to optimize thermal storage control.

[0036] As a preferred embodiment of the aquifer thermal storage peak-shaving method utilizing wind power curtailment described in this application, the calculation steps for the hourly load forecast specifically include:

[0037] The input parameters of the LSTM model include historical load data from the user side, real-time meteorological data, and holiday factors. Based on the input parameters, an LSTM model with an attention mechanism is used for sliding window training to obtain hourly load forecasts for the next day.

[0038] The LSTM model incorporating the attention mechanism specifically involves adding an attention weight layer to the hidden layer of the LSTM model, assigning different attention weights to different input parameters.

[0039] The system uses a preset first duration as a sliding window and uses historical user load data within the sliding window as training samples for training. The sliding window slides with a preset second duration as a step size, and the training samples are updated once each time it slides.

[0040] As a preferred embodiment of the aquifer thermal storage peak-shaving method utilizing wind power curtailment described in this application, the peak-shaving control specifically includes:

[0041] When the power grid peak-shaving command is received, the peak-shaving type is determined; the peak-shaving type includes thermal storage peak-shaving and thermal release peak-shaving.

[0042] If the power grid peak-shaving command is thermal storage peak-shaving, then obtain the real-time maximum power curtailment; multiply the real-time maximum power curtailment by the electrothermal conversion efficiency to obtain the target thermal storage power;

[0043] Calculate the maximum thermal storage power of the aquifer based on the rated power of the electric heating device and the heat dissipation capacity of the aquifer.

[0044] If the target thermal storage power is less than the maximum thermal storage power of the aquifer, a PID algorithm is used to adjust the actual thermal storage power to the target thermal storage power by controlling the electric heating device.

[0045] Otherwise, the difference between the maximum thermal storage power of the aquifer and the preset safety threshold is calculated, that is, the maximum thermal storage power of the aquifer minus the preset safety threshold, and the actual thermal storage power is adjusted to the difference by controlling the electric heating device using a PID algorithm.

[0046] As a preferred embodiment of the aquifer thermal storage peak-shaving method utilizing wind power curtailment described in this application, the peak-shaving control specifically further includes:

[0047] If the power grid peak-shaving command is heat release peak-shaving, the maximum heat release power of the aquifer is determined based on the rated power of the heat release device and the maximum transmission capacity of the user-side heating network.

[0048] If the maximum heat release power of the aquifer is greater than the load gap, a PID algorithm is used to adjust the actual heat release power to the load gap by controlling the flow rate and temperature of the heat exchanger.

[0049] Otherwise, calculate the difference between the maximum heat release power of the aquifer and the safety threshold, i.e., the maximum heat release power of the aquifer minus the safety threshold, and use a PID algorithm to adjust the actual heat release power to the difference by controlling the flow rate and temperature of the heat exchanger.

[0050] As a preferred embodiment of the aquifer thermal storage peak-shaving method for utilizing wind power curtailment described in this application, the results of the peak-shaving control include wind power curtailment absorption rate, aquifer thermal storage efficiency, grid peak-shaving command fulfillment rate, and user-side load supply rate.

[0051] The iteration termination conditions include: if the wind power curtailment absorption rate is greater than or equal to a preset absorption rate threshold, and the aquifer thermal storage efficiency is greater than or equal to a preset thermal storage efficiency threshold, and the grid peak-shaving command fulfillment rate is greater than or equal to a preset command fulfillment rate threshold, and the user-side load supply rate is greater than or equal to a preset load supply rate threshold, then the iteration termination conditions are met; if the number of iterations is greater than or equal to a preset maximum number of iterations, then the iteration meets the iteration termination conditions.

[0052] During each iteration, the actual heat loss and predicted heat loss of the previous iteration are calculated to obtain the heat loss error, and the peak power error of the previous iteration is also calculated. The compensation coefficient is adjusted in real time based on the heat loss error, and the parameters of the PID algorithm are adjusted based on the peak power error.

[0053] Secondly, this application provides an aquifer thermal storage peak-shaving system utilizing wind power curtailment, including a data acquisition module, a load forecasting module, a thermal storage peak-shaving module, and a feedback optimization module;

[0054] The data acquisition module is used to acquire wind power curtailment data and aquifer status data in real time and to preprocess the wind power curtailment data and aquifer status data.

[0055] Based on the wind power curtailment data, the load forecasting module uses an LSTM model with an attention mechanism to calculate hourly load forecasts and updates the LSTM model in real time.

[0056] The thermal energy storage peak shaving module is used to extract real-time power curtailment and set power curtailment thresholds; based on the real-time power curtailment and power curtailment thresholds, it makes matching decisions on power curtailment, thermal energy storage, and load supply; based on the aquifer state data and hourly load forecast values, and combined with the results of the matching decisions, it performs thermal energy storage control and heat release control.

[0057] The thermal energy storage peak shaving module is also used to receive grid peak shaving commands; based on the peak shaving commands and the results of thermal energy storage control and heat release control, it performs adaptive peak shaving control and calculates the results of peak shaving control.

[0058] Based on the results of the peak-shaving control, the feedback optimization module iteratively executes the method of the thermal storage peak-shaving module; when the preset iteration conditions are met, the iteration stops and the peak-shaving control result obtained from the last iteration is output.

[0059] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0060] Based on the acquired wind power curtailment data, hourly load forecasts are calculated, and real-time curtailment volume is extracted. Based on the real-time curtailment volume and a preset curtailment threshold, matching decisions are made regarding curtailment, thermal storage, and load supply. Based on the acquired aquifer state data and hourly load forecasts, and combined with the matching decision results, thermal storage and heat release control are implemented, improving the curtailment absorption rate and aquifer thermal storage efficiency. This achieves efficient conversion and storage of wind power curtailment without relying on high-cost equipment such as thermal power plants and gas turbines, and also overcomes the scale limitations of electrochemical energy storage. Adaptive peak-shaving control is performed based on peak-shaving commands and the results of the thermal storage and heat release control, and the peak-shaving control results are calculated. Iterative matching decisions and peak-shaving control are then performed based on the peak-shaving control results, improving the grid peak-shaving command fulfillment rate and achieving a dual improvement in grid peak-shaving accuracy and user load stability, meeting the safety and stability requirements of large-scale, long-duration peak-shaving scenarios. Attached Figure Description

[0061] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0062] Figure 1 A flowchart of a method for peak shaving using aquifer thermal storage based on wind power curtailment provided in this application;

[0063] Figure 2 This application provides a structural diagram of an aquifer thermal storage and peak-shaving system that utilizes abandoned wind power. Detailed Implementation

[0064] The technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments and specific features in the embodiments are detailed descriptions of the technical solution of this application, rather than limitations thereof. In the absence of conflict, the embodiments and technical features in the embodiments can be combined with each other.

[0065] Example 1

[0066] like Figure 1 As shown in the figure, this embodiment introduces a method for peak shaving using aquifer thermal storage based on wind power curtailment, including:

[0067] S1: Real-time synchronous acquisition of wind power curtailment data and aquifer status data, and preprocessing of the wind power curtailment data and aquifer status data; the wind power curtailment data includes real-time wind power data, load data and grid data; the aquifer status data includes aquifer parameters and real-time data of the thermal storage area;

[0068] Specifically, the real-time wind power data includes real-time curtailment, wind power installed capacity, wind power output fluctuation range, and curtailment duration; the load data includes user-side historical load data, real-time meteorological data, and holiday factors; the user-side historical load data is hourly load data for the past N months, where N is a preset number of months; the real-time meteorological data includes temperature, precipitation, and wind speed; the power grid data includes power grid peak-shaving instructions, power grid conventional heating power, power grid frequency target range, and maximum allowable load fluctuation value of the power grid.

[0069] In this embodiment, the parameters of the aquifer are obtained through geological exploration, including the aquifer thermal conductivity, porosity, effective thickness of the reservoir, permeability, and underground ambient temperature. The real-time data of the reservoir is obtained by deploying temperature sensors, pressure sensors, and capacity monitoring devices in the reservoir area of ​​the aquifer to collect the reservoir temperature, reservoir pressure, and remaining reservoir capacity in real time. The deployment density of the sensors is determined according to the area of ​​the reservoir, with at least 3 sets of sensors deployed per 1000m² to ensure that the real-time data of the reservoir covers the entire reservoir area, and the sensing error is controlled by β to ensure data accuracy.

[0070] The specific method for preprocessing the wind power curtailment data and aquifer status data is as follows: Kalman filtering is applied to the real-time wind power data to eliminate random noise; outlier detection and removal are performed on the aquifer status data to filter out interference; and the units and formats of the load data and grid data are standardized to ensure data format consistency.

[0071] S2: Based on the wind power curtailment data, an LSTM (Long Short-Term Memory) network model incorporating an attention mechanism is used to calculate hourly load forecasts and update the LSTM model in real time. Specifically, the LSTM model incorporating an attention mechanism adds an attention weight layer to the hidden layers of the LSTM model, assigning different attention weights to different input parameters to focus on key influencing factors and reduce interference from irrelevant features. For example, the attention weight of the load is increased by 0.1 for every 1°C drop in winter temperature, and the attention weight of the load is adjusted to 1.2 times that of weekdays during holidays.

[0072] The hourly load forecast is calculated as follows: based on the input parameters, an LSTM model with an attention mechanism is used for sliding window training to obtain the hourly load forecast for the next day; the LSTM model with an attention mechanism not only improves the prediction accuracy of long sequences, but also reduces the prediction error.

[0073] The input parameters of the LSTM model include historical load data from the user side, real-time meteorological data, and holiday factors. The historical load data from the user side includes categorized load data such as residential load, commercial load, and industrial load. This historical load data is stored at an hourly granularity, and data from abnormal periods such as power outages and equipment failures is excluded to ensure data validity. The real-time meteorological data includes temperature, relative humidity, wind speed, and sunshine duration. This real-time meteorological data is obtained from local meteorological stations or meteorological data platforms and updated at a preset frequency. The holiday factors are quantized using binary quantization.

[0074] Specifically, holidays are set to 1, and working days are set to 0; holidays include weekends; if the holidays are special holidays, such as Spring Festival and National Day, the holiday factor value is further refined, and special holidays are set to 2 to reflect the load differences of different holidays.

[0075] The specific method of the sliding window training is as follows: a preset first duration is used as the sliding window, and historical load data of the user side within the sliding window is used as training samples for training; the sliding window slides with a preset second duration as the step size, and the training samples are updated once each slide; in this embodiment, the preset first duration is 24 hours, and the preset second duration is 1 hour, to adapt to the daily cycle variation of the load; for example, for the characteristics of the morning and evening peak loads of residents, the sliding window training can enhance the prediction accuracy of the load during peak hours and stabilize the prediction error.

[0076] The LSTM model is updated as follows: A preset third time interval is used as the update interval. The real-time collected load data is compared with the hourly load forecast for the corresponding time period, and the root mean square error (RMSE) between the load data and the hourly load forecast for the corresponding time period is calculated. Based on the RMSE, the attention weights of the LSTM model are adjusted using a backpropagation algorithm. For example, when the RMSE between the actual load data and the hourly load forecast for the corresponding time period is greater than a preset error threshold, the attention weights of the corresponding meteorological data are increased to ensure that the LSTM model dynamically optimizes with load changes.

[0077] S3: Extract the real-time abandoned power and set the abandoned power threshold; based on the real-time abandoned power and the abandoned power threshold, make matching decisions on abandoned power, thermal storage and load supply;

[0078] Specifically, based on the minimum stable operating load of the power grid and the maximum daily thermal storage capacity of the aquifer, the power curtailment threshold is determined and the real-time power curtailment is extracted. For example, if the minimum stable load of the power grid in a certain area is 20MW and the maximum daily thermal storage capacity of the aquifer corresponds to a power curtailment of 15MW, then the power curtailment threshold is set to 5MW to ensure that the portion of the power curtailment exceeding the stable load of the power grid can be absorbed by thermal storage.

[0079] The specific method of the matching decision is as follows: if the real-time abandoned power is greater than or equal to the abandoned power threshold, then abandoned power consumption and thermal storage are carried out; if the real-time abandoned power is less than the abandoned power threshold, then thermal storage and load supply are carried out.

[0080] S4: Based on the aquifer state data and hourly load forecasts, and combined with the matching decision results, thermal storage control and heat release control are performed; if the matching result is power curtailment and thermal storage, then thermal storage control is performed; the specific method of thermal storage control is as follows:

[0081] The underground ambient temperature is obtained, and the temperature difference between the underground ambient temperature and the temperature of the thermal storage area is calculated, i.e., the underground ambient temperature minus the temperature of the thermal storage area; the thermal conductivity of the aquifer is calculated by measuring the aquifer core sample in the laboratory, and the thermal conductivity of the aquifer is multiplied by the dimensionless temperature difference to obtain the first product;

[0082] In this embodiment, the underground ambient temperature is obtained by drilling holes around the thermal storage area to deploy depth sensors and monitoring the temperature of the underground constant temperature layer for at least 30 consecutive days to obtain monitoring data; the average value of the monitoring data is calculated to obtain the underground ambient temperature.

[0083] Calculate the effective thickness of the thermal storage zone and obtain the porosity; multiply the effective thickness of the thermal storage zone and the porosity after removing the dimensions to obtain a second product; the effective thickness of the thermal storage zone is calculated as follows: by ground-penetrating radar exploration or core drilling analysis, the effective thickness of the thermal storage medium distribution is determined, and the thickness of rock layers without thermal storage medium is excluded to obtain the effective thickness of the thermal storage zone;

[0084] The heat loss rate is obtained by calculating the ratio of the first product to the second product. The heat loss rate reflects the proportion of heat loss from the heat storage area to the surrounding environment per unit time. When the temperature of the heat storage area is greater than the temperature of the underground environment, the heat loss rate is greater than 0.

[0085] The thermal storage power is calculated based on the real-time abandoned power, and the heat loss compensation power is calculated based on the thermal storage power.

[0086] Specifically, the remaining thermal storage capacity, specific heat capacity of the thermal storage medium, and allowable temperature rise of the thermal storage medium are obtained; the thermal storage power is less than or equal to the cumulative product of the remaining thermal storage capacity, specific heat capacity of the thermal storage medium, and allowable temperature rise of the thermal storage medium; the thermal storage power refers to the power input from the electric heating device to the aquifer, and its calculation method is as follows: determine the electrothermal conversion efficiency according to the type of electric heating device, and multiply the real-time abandoned power by the electrothermal conversion efficiency to obtain the thermal storage power;

[0087] The thermal compensation power is calculated as follows: obtain the historical heat loss compensation effect, and set the compensation coefficient according to the historical heat loss compensation effect; calculate the cumulative product of the heat loss rate, the thermal storage power and the compensation coefficient to obtain the thermal compensation power;

[0088] Specifically, the temperature fluctuation value of the thermal storage area is calculated, and a compensation coefficient is set based on the temperature fluctuation value; the compensation coefficient is positively correlated with the temperature fluctuation value to ensure that the compensation power covers the actual heat loss.

[0089] The initial thermal storage power is obtained, and the thermal compensation power is added to the initial thermal storage power to obtain the actual thermal storage power; the output current or voltage of the electrothermal conversion device is adjusted to control the thermal storage area to store heat with the actual thermal storage power, so as to improve the thermal storage efficiency of the aquifer thermal storage area; the adjustment process is smooth to avoid the impact of power change on the power grid or thermal storage device.

[0090] If the matching result is heat storage and load supply, then heat release control is performed; the specific method of heat release control is as follows:

[0091] The current heat storage capacity is calculated based on the temperature, volume, and specific heat capacity of the heat storage medium in the heat storage area, and the remaining heat storage capacity is calculated based on the current heat storage capacity; the remaining heat storage capacity is the difference between the remaining heat storage capacity and the current heat storage capacity, that is, the remaining heat storage capacity minus the current heat storage capacity.

[0092] Obtain the conventional heating power of the power grid, and calculate the load gap based on the conventional heating power of the power grid and the hourly load forecast; the load gap is the difference between the hourly load forecast and the conventional heating power of the power grid, that is, the hourly load forecast minus the conventional heating power of the power grid.

[0093] If the thermal storage capacity is greater than or equal to the load gap, the thermal storage capacity is converted into actual heat release power through the heat release device of the aquifer to supplement the load gap; the actual heat release power is dynamically adjusted according to the load gap to ensure supply and demand balance.

[0094] If the thermal storage capacity is less than the load gap, an auxiliary heating request is sent to the power grid dispatch center to call on the heating reserve capacity of conventional power sources such as thermal power units and gas turbines to fill the remaining load gap. At the same time, the load gap data is recorded for subsequent optimization of thermal storage strategies, such as raising the curtailment threshold or increasing thermal storage capacity.

[0095] S5: Receive the power grid peak shaving command; perform adaptive peak shaving control based on the peak shaving command and the results of the thermal storage control and heat release control, and calculate the result of the peak shaving control;

[0096] Upon receiving the power grid peak-shaving command, the peak-shaving type is determined; the peak-shaving type includes thermal storage peak-shaving and thermal release peak-shaving; the peak-shaving control specifically includes:

[0097] If the power grid peak-shaving command is thermal storage peak-shaving, then obtain the real-time maximum power curtailment; multiply the real-time maximum power curtailment by the electrothermal conversion efficiency to obtain the target thermal storage power;

[0098] The maximum thermal storage power of the aquifer is calculated based on the rated power of the electric heating device and the heat dissipation capacity of the aquifer. If the target thermal storage power is less than the maximum thermal storage power of the aquifer, a PID algorithm is used to adjust the actual thermal storage power to the target thermal storage power by controlling the electric heating device. Otherwise, the difference between the maximum thermal storage power of the aquifer and a preset safety threshold is calculated, i.e., the maximum thermal storage power of the aquifer minus the preset safety threshold, and a PID algorithm is used to adjust the actual thermal storage power to the difference by controlling the electric heating device.

[0099] If the power grid peak-shaving command is heat release peak-shaving, the maximum heat release power of the aquifer is determined based on the rated power of the heat release device and the maximum transmission capacity of the user-side heating network. If the maximum heat release power of the aquifer is greater than the load gap, a PID algorithm is used to adjust the actual heat release power to the load gap by regulating the flow rate and temperature of the heat exchanger. Otherwise, the difference between the maximum heat release power of the aquifer and the safety threshold is calculated, i.e., the maximum heat release power of the aquifer minus the safety threshold, and a PID algorithm is used to adjust the actual heat release power to the difference by regulating the flow rate and temperature of the heat exchanger.

[0100] Optionally, a peak-shaving response time parameter can be set. Regardless of the peak-shaving type, the response speed must be controlled by the peak-shaving response time parameter. The time from receiving the peak-shaving command to the power adjustment being completed should not exceed the peak-shaving response time parameter to avoid delaying the grid's peak-shaving opportunity. At the same time, during the adjustment process, the changes in actual thermal storage power or actual thermal release power are monitored in real time. If the actual thermal storage power exceeds the preset ratio of the target thermal storage power or the actual thermal release power exceeds the preset ratio of the target thermal release power, fine-tuning is triggered immediately to ensure peak-shaving accuracy.

[0101] The results of the peak shaving control include wind power curtailment rate, aquifer thermal storage efficiency, grid peak shaving command fulfillment rate, and user-side load supply rate. The wind power curtailment rate reflects the utilization efficiency of curtailed power resources. It is calculated by obtaining the amount of curtailed power consumed and taking the ratio of the consumed curtailed power to the total curtailed power as the wind power curtailment rate.

[0102] The aquifer thermal storage efficiency is calculated as follows: the actual heat storage capacity of the storage area is obtained, and the ratio of the actual heat storage capacity to the theoretical maximum heat storage capacity is taken as the aquifer thermal storage efficiency; the theoretical maximum heat storage capacity is determined by the volume of the thermal storage medium, the specific heat capacity of the thermal storage medium, and the allowable temperature rise of the thermal storage medium.

[0103] The grid peak-shaving command satisfaction rate reflects the system's responsiveness to grid peak-shaving demands. It is calculated by obtaining the number of times peak-shaving requirements are met and the total number of peak-shaving commands, and using the ratio of the number of times peak-shaving requirements are met to the total number of peak-shaving commands as the grid peak-shaving command satisfaction rate.

[0104] The user-side load supply rate reflects the degree to which load demand is met. It is calculated by calculating the actual heat supply and taking the ratio of the actual heat supply to the total load demand on the user side as the user-side load supply rate.

[0105] S6: Based on the results of the peak shaving control, iteratively execute steps S3 to S5; when the preset iteration conditions are met, stop the iteration and output the peak shaving control results obtained in the last iteration; the iteration conditions include: the results of the peak shaving control are all greater than or equal to the preset threshold, or the number of iterations is greater than or equal to the preset maximum number of iterations;

[0106] Specifically, if the wind power curtailment absorption rate is greater than or equal to a preset absorption rate threshold, and the aquifer thermal storage efficiency is greater than or equal to a preset thermal storage efficiency threshold, and the grid peak shaving command fulfillment rate is greater than or equal to a preset command fulfillment rate threshold, and the user-side load supply rate is greater than or equal to a preset load supply rate threshold, then the iteration meets the iteration termination condition; if the number of iterations is greater than or equal to a preset maximum number of iterations, then the iteration meets the iteration termination condition; otherwise, based on the results of the peak shaving control, steps S3 to S5 are iteratively executed.

[0107] In this embodiment, during each iteration, the actual heat loss and predicted heat loss of the previous iteration are calculated to obtain the heat loss error, and the peak power error of the previous iteration is also calculated. The compensation coefficient is adjusted in real time based on the heat loss error, and the parameters of the PID algorithm are adjusted based on the peak power error. For example, if the peak power overshoots, the proportional coefficient is reduced to improve system stability.

[0108] Optionally, the results of the peak shaving control can be displayed in real time on a visual dashboard, and data export and historical data backtracking can be supported.

[0109] Example 2

[0110] This embodiment is the second embodiment of this application; it is based on the same inventive concept as Embodiment 1, and refers to... Figure 2 This embodiment introduces an aquifer thermal storage peak-shaving system utilizing wind power curtailment, including a data acquisition module, a load forecasting module, a thermal storage peak-shaving module, and a feedback optimization module;

[0111] The data acquisition module is used to acquire wind power curtailment data and aquifer status data in real time and to preprocess the wind power curtailment data and aquifer status data.

[0112] Based on the wind power curtailment data, the load forecasting module uses an LSTM model with an attention mechanism to calculate hourly load forecasts and updates the LSTM model in real time.

[0113] The thermal energy storage peak shaving module is used to extract real-time power curtailment and set power curtailment thresholds; based on the real-time power curtailment and power curtailment thresholds, it makes matching decisions on power curtailment, thermal energy storage, and load supply; based on the aquifer state data and hourly load forecast values, and combined with the results of the matching decisions, it performs thermal energy storage control and heat release control.

[0114] The thermal energy storage peak shaving module is also used to receive grid peak shaving commands; based on the peak shaving commands and the results of thermal energy storage control and heat release control, it performs adaptive peak shaving control and calculates the results of peak shaving control.

[0115] Based on the results of the peak-shaving control, the feedback optimization module iteratively executes all steps of the thermal storage peak-shaving module; when the preset iteration conditions are met, the iteration stops and the peak-shaving control result obtained from the last iteration is output.

[0116] The specific functions of each of the above steps are described in the relevant content of the method for peak shaving by utilizing wind power curtailment in aquifer thermal storage as described in Example 1, and will not be repeated here.

[0117] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0118] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of protection of this application, and these forms are all within the protection scope of this application.

Claims

1. An aquifer thermal energy storage peak shaving method using wind power curtailment, characterized in that, The method comprises the following steps: S1: Real-time synchronous acquisition of wind power curtailment data and aquifer state data, and preprocessing of the wind power curtailment data and aquifer state data; S2: Based on the wind power curtailment data, an LSTM model with attention mechanism is used to calculate the hourly load prediction value, and the LSTM model is updated in real time; S3: Extracting real-time curtailment amount and setting curtailment threshold; based on the real-time curtailment amount and the curtailment threshold, making matching decisions of curtailment, heat storage and load supply; S4: Based on the aquifer state data and the hourly load prediction value, and combining the results of the matching decisions, heat storage control and heat release control are carried out; S5: Receiving grid peak shaving instructions; according to the peak shaving instructions and the results of the heat storage control and the heat release control, adaptive peak shaving control is carried out, and the results of the peak shaving control are calculated; S6: Based on the results of the peak shaving control, steps S3 to S5 are repeatedly executed; when the preset iteration condition is met, the iteration is stopped, and the results of the peak shaving control obtained in the last iteration are output.

2. The aquifer thermal energy storage peak shaving method using wind power curtailment of claim 1, wherein, The wind power curtailment data includes real-time curtailment amount; the step of heat storage control specifically comprises: Obtaining the underground environment temperature and calculating the temperature difference between the underground environment temperature and the heat storage area temperature, i.e. the underground environment temperature minus the heat storage area temperature; Calculating the thermal conductivity of the aquifer by laboratory determination of the aquifer core sample, and multiplying the thermal conductivity of the aquifer and the temperature difference after removing the dimension to obtain a first product; Calculating the effective thickness of the heat storage area and obtaining the porosity; multiplying the effective thickness of the heat storage area and the porosity after removing the dimension to obtain a second product; Calculating the ratio of the first product and the second product to obtain the heat loss rate; According to the type of electric heating device, the electric heating conversion efficiency is determined, the real-time curtailment amount and the electric heating conversion efficiency are multiplied to obtain the heat storage power.

3. The aquifer thermal energy storage peak shaving method using wind power curtailment of claim 2, wherein, The step of heat storage control further comprises: Obtaining the residual heat storage capacity of the heat storage area, the specific heat capacity of the heat storage medium and the allowable temperature rise of the heat storage medium; the heat storage power is less than or equal to the cumulative product of the residual heat storage capacity, the specific heat capacity of the heat storage medium and the allowable temperature rise of the heat storage medium; Calculating the temperature fluctuation value of the heat storage area, and setting a compensation coefficient based on the temperature fluctuation value; the compensation coefficient is positively correlated with the temperature fluctuation value; Calculating the cumulative product of the heat loss rate, the heat storage power and the compensation coefficient to obtain the heat compensation power; obtaining the initial heat storage power, and adding the heat compensation power and the initial heat storage power to obtain the actual heat storage power; Adjusting the output current or voltage of the electric heating conversion device to control the heat storage area to store heat at the actual heat storage power.

4. The aquifer thermal energy storage peak shaving method using wind power curtailment of claim 3, wherein, The step of heat release control specifically comprises: Based on the temperature, volume and specific heat capacity of the heat storage medium, the current heat storage amount is calculated, and the heat storage surplus is calculated based on the current heat storage amount; The heat storage surplus is the difference between the residual heat storage surplus and the current heat storage amount, i.e. the residual heat storage surplus minus the current heat storage amount; Obtaining the grid regular heating power, and calculating the load gap based on the grid regular heating power and the hourly load prediction value; The load gap is the difference between the hourly load prediction value and the grid regular heating power, i.e. the hourly load prediction value minus the grid regular heating power.

5. The aquifer thermal energy storage peak shaving method using wind power curtailment of claim 4, wherein, The step of heat release control further comprises: If the heat storage surplus is greater than or equal to the load gap, the heat storage surplus is converted into actual heat release power by a heat release device of the aquifer; The actual heat release power is dynamically adjusted according to the load gap; If the heat storage surplus is less than the load gap, an auxiliary heat supply request is sent to a power grid dispatching center, standby heat supply capacity of conventional power sources such as thermal power generating units and gas turbines is called, and the remaining load gap is filled; Load gap data is recorded for optimizing heat storage control.

6. The aquifer thermal energy storage peak shaving method using wind power curtailment of claim 5, wherein, The calculation of the hourly load prediction value specifically includes: The input parameters of the LSTM model include historical user-side load data, real-time weather data, and holiday factors; based on the input parameters, an LSTM model with an attention mechanism is used for sliding window training to obtain the hourly load prediction value for the next day; The LSTM model with the attention mechanism specifically includes: an attention weight layer is added to the hidden layer of the LSTM model, and different attention weights are given to different input parameters; A preset first length is used as a sliding window, and historical user-side load data in the sliding window is used as a training sample for training; the sliding window slides at a preset second length as a step, and the training sample is updated once every time the sliding window slides.

7. The aquifer thermal energy storage peak shaving method using wind power curtailment of claim 6, wherein, The peak regulation control specifically includes: When the grid peak regulation instruction is received, the type of peak regulation is determined; the type of peak regulation includes heat storage peak regulation and heat release peak regulation; If the grid peak regulation instruction is heat storage peak regulation, the real-time maximum curtailed power is obtained; the real-time maximum curtailed power is multiplied by the electric-thermal conversion efficiency to obtain a target heat storage power; The maximum heat storage power of the aquifer is calculated based on the rated power of the electric heating device and the heat dissipation capacity of the aquifer; If the target heat storage power is less than the maximum heat storage power of the aquifer, a PID algorithm is used to adjust the actual heat storage power to the target heat storage power by regulating the electric heating device; Otherwise, the difference between the maximum heat storage power of the aquifer and the preset safety threshold, i.e., the maximum heat storage power of the aquifer minus the preset safety threshold, is calculated, and a PID algorithm is used to adjust the actual heat storage power to the difference by regulating the electric heating device. 8.The method of claim 7, wherein, The peak regulation control specifically further includes: If the grid peak regulation instruction is heat release peak regulation, the maximum heat release power of the aquifer is determined according to the rated power of the heat release device and the maximum delivery capacity of the user-side heat supply pipe network; If the maximum heat release power of the aquifer is greater than the load gap, a PID algorithm is used to adjust the actual heat release power to the load gap by regulating the flow and temperature of the heat exchanger; Otherwise, the difference between the maximum heat release power of the aquifer and the safety threshold, i.e., the maximum heat release power of the aquifer minus the safety threshold, is calculated, and a PID algorithm is used to adjust the actual heat release power to the difference by regulating the flow and temperature of the heat exchanger. 9.The method of claim 8, wherein, The results of the peak regulation control include the wind power curtailment consumption rate, the aquifer heat storage efficiency, the grid peak regulation instruction satisfaction rate, and the user-side load supply rate. The iteration termination condition comprises: if the wind power curtailment accommodation rate is greater than or equal to a preset accommodation rate threshold, the aquifer heat storage efficiency is greater than or equal to a preset heat storage efficiency threshold, the power grid peak shaving instruction satisfaction rate is greater than or equal to a preset instruction satisfaction rate threshold, and the user side load supply rate is greater than or equal to a preset load supply rate threshold, the iteration termination condition is satisfied; if the iteration number is greater than or equal to a preset maximum iteration number, the iteration satisfies the iteration termination condition. In each iteration, the actual heat loss and the predicted heat loss of the previous iteration are calculated to obtain a heat loss error, and the peak shaving power error of the previous iteration is calculated; the compensation coefficient is adjusted in real time based on the heat loss error, and the parameters of the PID algorithm are adjusted based on the peak shaving power error.

10. An aquifer thermal energy storage peak shaving system for use in implementing a method of utilizing wind curtailment for thermal energy storage in an aquifer as claimed in any one of claims 1-9, characterized in that, The system comprises a data acquisition module, a load prediction module, a heat storage peak shaving module and a feedback optimization module. The data acquisition module is configured to acquire wind power curtailment data and aquifer state data in real time and synchronously, and preprocess the wind power curtailment data and the aquifer state data. The load prediction module is configured to calculate a per-hour load prediction value based on the wind power curtailment data by using an LSTM model with an attention mechanism, and update the LSTM model in real time. The heat storage peak shaving module is configured to extract real-time curtailment amount and set a curtailment threshold; based on the real-time curtailment amount and the curtailment threshold, make a matching decision on curtailment, heat storage and load supply; based on the aquifer state data and the per-hour load prediction value, and in combination with the result of the matching decision, make heat storage control and heat release control. The heat storage peak shaving module is further configured to receive a power grid peak shaving instruction; based on the peak shaving instruction and the results of the heat storage control and the heat release control, make adaptive peak shaving control, and calculate the result of the peak shaving control. The feedback optimization module is configured to iteratively execute the method of the heat storage peak shaving module based on the result of the peak shaving control; when a preset iteration condition is satisfied, stop iteration, and output the result of the peak shaving control obtained in the last iteration.

Images