Energy storage and consumption reduction control method, system and equipment and storage medium

Abstract

The invention relates to the technical field of energy storage, in particular to an energy storage consumption reduction control method, system and device and a storage medium, and the method comprises the steps: obtaining multi-source data and battery parameters; inputting the multi-source data into a pre-trained prediction model for prediction to obtain an environment temperature prediction value; evaluating the life cycle of the battery according to the battery parameters and setting a pre-cooling target temperature interval; and a pre-cooling operation plan is made according to the pre-cooling target temperature interval and the predicted environment temperature value, and the operation power is adjusted according to the pre-cooling operation plan, so that energy storage and consumption reduction of the battery are achieved. The auxiliary power loss is reduced, and the economical efficiency is improved.

Description

Technical Field

[0001] This disclosure relates to the field of energy storage technology, and in particular to an energy storage consumption reduction control method, system, device and storage medium. Background Technology

[0002] In the energy storage field, air-cooled and liquid-cooled energy storage systems are the mainstream temperature control solutions, with liquid-cooled units widely used due to their superior heat dissipation efficiency. However, existing liquid-cooled units are controlled by triggering operation through a set return water temperature threshold, and their control logic relies solely on this single parameter. For example, the heat generation power of a battery during standby is only 1 / 5 to 1 / 8 of that during charging, yet it still needs to maintain the same cooling intensity as during charging, resulting in significant wasted energy. Data shows that cooling system losses account for 70% of the total losses in a DC energy storage system, with 30% to 40% of these losses occurring during standby. Furthermore, existing control schemes do not incorporate dynamic adjustments based on peak and off-peak electricity prices, maintaining full-load operation during peak electricity price periods, further increasing auxiliary power costs. This leads to excessive auxiliary power losses and insufficient economic efficiency, becoming a core bottleneck restricting the large-scale application of energy storage systems. Summary of the Invention

[0003] To address the aforementioned technical problems, this disclosure provides an energy storage and consumption reduction control method, comprising: Acquire multi-source data and battery parameters; The multi-source data is input into a pre-trained prediction model to make predictions and obtain the predicted ambient temperature value. The target pre-cooling temperature range is set based on the battery parameters to assess the battery lifespan. A pre-cooling operation plan is formulated based on the pre-cooling target temperature range and the predicted ambient temperature, and the operating power is adjusted according to the pre-cooling operation plan to reduce energy consumption of battery storage.

[0004] Furthermore, the training of the prediction model includes: Use historical multi-source data as training samples; The training samples are input into the prediction model, and the prediction model is iterated using a regression loss function to obtain the performance index of the current model training. When the number of iterations in which the performance metric has not improved reaches a preset value, the model parameters of the current model are saved to obtain the trained prediction model.

[0005] Furthermore, before the step of formulating a pre-cooling operation plan based on the pre-cooling target temperature range and the predicted ambient temperature, and adjusting the operating power according to the pre-cooling operation plan to reduce battery energy consumption, the method further includes: Plot the temperature decay curve based on the current battery characteristics; Based on the temperature decay curve, the battery pre-cooling time is obtained; The precooling operation plan is formulated based on the precooling time, the target precooling temperature range, and the predicted ambient temperature.

[0006] Furthermore, the step of formulating a pre-cooling operation plan based on the pre-cooling target temperature range and the predicted ambient temperature, and adjusting the operating power according to the pre-cooling operation plan to reduce battery energy consumption, includes: Obtain electricity price information; During periods of low electricity prices, the operating power is adjusted to a preset first range; During the period of stable electricity prices, the operating power is adjusted to a preset second range; During peak electricity price periods, the operating power will be adjusted to a preset third range.

[0007] Furthermore, during the period of low electricity prices, the temperature will be reduced to at least one degree below the lower limit of the pre-cooling target temperature range.

[0008] Furthermore, when the predicted ambient temperature is less than the preset temperature, precooling is turned off.

[0009] Furthermore, the multi-source data includes weather forecasts, historical ambient temperature data for the same period, and temperature transmission delay data from the energy storage container.

[0010] This disclosure also provides an energy storage and consumption reduction control system, including: The acquisition module is used to acquire multi-source data and battery parameters; The prediction module is used to input the multi-source data into a pre-trained prediction model to make predictions and obtain the predicted ambient temperature value. The setting module is used to set the target pre-cooling temperature range based on battery parameters to assess the battery lifespan. The adjustment module is used to formulate a pre-cooling operation plan based on the pre-cooling target temperature range and the predicted ambient temperature, and adjust the operating power according to the pre-cooling operation plan to reduce energy consumption of battery storage.

[0011] This disclosure also provides a computer device, including a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the energy storage and consumption reduction control method.

[0012] This disclosure also provides a computer-readable storage medium storing computer-readable instructions, which, when executed by a processor, implement the steps of the energy storage and power consumption reduction control method.

[0013] The technical solution provided in this disclosure has the following advantages compared with the prior art: This disclosure achieves a reduction in cooling energy consumption under typical operating conditions by implementing quantitative pre-cooling during off-peak electricity price periods based on ambient temperature prediction, and significantly reducing or stopping active cooling during peak periods by utilizing temperature decay. It directly couples the target pre-cooling temperature range with the battery cycle life, avoiding the surge in internal resistance caused by low temperatures and the material dissolution caused by high temperatures, thus maintaining a low capacity decay rate. This results in reduced auxiliary power losses and improved economic efficiency. Attached Figure Description

[0014] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0015] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the energy storage and consumption reduction control method described in the embodiments of this disclosure; Figure 2 This is an embodiment of the present disclosure. Figure 1 A schematic diagram of the method for adjusting operating power as described above; Figure 3 This is a schematic diagram of the energy storage and consumption reduction control system described in an embodiment of this disclosure. Detailed Implementation

[0017] To better understand the above-mentioned objectives, features, and advantages of this disclosure, the solutions disclosed herein will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0018] Numerous specific details are set forth in the following description in order to provide a full understanding of this disclosure, but this disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some, and not all, of the embodiments of this disclosure.

[0019] Figure 1 This is a schematic diagram of the energy storage and energy consumption reduction control method described in the embodiments of this disclosure; as follows: Figure 1 As shown, an energy storage consumption reduction control method includes: Step S1: Obtain multi-source data and battery parameters; In this embodiment, multi-source data is collected (weather forecasts, historical ambient temperature data, and temperature conduction delay data of the energy storage container; internal temperature sequences and container humidity can also be collected). Real-time operational data (cell or pack temperature, internal container temperature, ambient temperature, humidity, voltage, current, SOC, cooling equipment status and power level) can also be collected as needed. Simultaneously, battery parameters (battery chemistry type, cycle count, and cycle life, etc.) are read from the battery management system or asset database. After data acquisition, preprocessing is performed, including time alignment, missing value imputation, outlier detection and removal, and filtering (low-pass or smoothing). Additionally, the uncertainty index of the weather forecast can be calculated as a confidence input for subsequent decision-making. By establishing a complete, time-consistent, and quality-controllable data foundation, accurate and reliable inputs are provided for subsequent ambient temperature prediction and battery life-cycle-based strategies. By acquiring and cleaning historical data and conduction delay data in advance, prediction accuracy can be significantly improved and the risk of scheduling misjudgment can be reduced, thereby improving the economy and operational safety of off-peak pre-cooling.

[0020] Step S2: Input the multi-source data into the pre-trained prediction model to make predictions and obtain the predicted ambient temperature value; In this embodiment, preprocessed time-series features (including hourly weather forecasts, historical temperature lag sequences within the chamber, chamber humidity, wind speed, planned charge / discharge power trajectories, SOC, PACK type, etc.) are fed into a pre-trained prediction model (i.e., a neural network model). The model training phase employs a loss and early shutdown strategy matched to the task. During online prediction, the model simultaneously outputs hourly ambient temperature predictions for the next H hours (e.g., 6 hours). This provides hourly, quantifiable confidence-based ambient temperature time-series forecasts, enabling the scheduler to make power allocation and overcooling decisions in advance. High-quality ambient temperature predictions can significantly reduce overcooling or undercooling caused by sudden environmental changes, thereby improving the accuracy of peak-shifting energy saving and reducing the negative impact on battery life, reducing unnecessary cooling energy consumption, and avoiding accelerated aging caused by temperature fluctuations.

[0021] Step S3: Based on the battery parameters, assess the battery life cycle and set the pre-cooling target temperature range; In this embodiment, based on the read battery parameters, especially the battery cycle count or equivalent lifespan classification, the system uses lifetime and temperature empirical models or experimental fitting curves to determine the appropriate pre-cooling target temperature range for the battery classification. For example, for lithium iron phosphate batteries with more than 5000 cycles, the system sets the target range to 22 to 26 °C and uses this range as a constraint for subsequent scheduling; for other cycle counts or chemical systems, the corresponding range is remapped according to offline lifetime test data. By converting the rules into executable engineering thresholds, the pre-cooling and charging actions are directly coupled with battery lifespan, thereby minimizing energy consumption while ensuring battery safety and lifespan. For example, using the 22 to 26 °C range under conditions of more than 5000 cycles has been verified to suppress the increase in low-temperature internal resistance and control the capacity decay rate to within 1.0% / 100 cycles, improving the long-term economy and reliability of the energy storage system.

[0022] Step S4: Develop a pre-cooling operation plan based on the pre-cooling target temperature range and the predicted ambient temperature, and adjust the operating power according to the pre-cooling operation plan to reduce energy consumption of battery storage.

[0023] In this embodiment, the temperature decay model is invoked and combined with the ambient temperature prediction value to calculate the sustainable time (i.e., the time required from pre-cooling to recovery to the target upper limit) under different excess cooling depths. Then, based on electricity price information, energy balance model (cold energy conservation and cold energy to temperature conversion coefficient), equipment power upper and lower limits, and safety constraints, a multi-period optimization problem is constructed, and model predictive control or a fast solver is used to obtain the cooling power scheduling curve for future periods. Strategically, high-power pre-cooling is performed during periods of low electricity prices (first interval, example: 80% to 100% of rated power) and excess cooling can be performed (i.e., the temperature is reduced to 1 to 2°C below the lower limit of the pre-cooling target temperature range) to reserve cold energy. During periods of flat and peak electricity prices, the power is reduced to 30% to 50% and 10% to 20% of the rated power, respectively, to achieve peak shifting. At the same time, after pre-cooling, the system enters the cold preservation phase and is shut off or reduced to low power. Low-power supplemental cooling (20% to 30%) is only used when the temperature inside the box approaches the upper limit. By translating predictions and lifetime constraints into specific power timing execution schemes, this approach fully leverages electricity price differences to achieve economic efficiency (e.g., off-peak electricity price of 0.3 yuan / kWh, peak price of 1.2 yuan / kWh, and excess cooling of 1°C can replace 0.8 kWh / h of peak-hour cooling, thus saving on electricity costs). It also utilizes temperature inertia (e.g., liquid cooling holding time of 8 to 10 hours, air cooling of 4 to 5 hours) to reduce active cooling time and energy consumption. Overall, this approach can reduce cooling energy consumption by approximately 25% to 40%, peak-hour maintenance power by 60% to 70%, and standby energy consumption by approximately 80%, while extending battery life and reducing operating costs while meeting safety thresholds.

[0024] This disclosure achieves a reduction in cooling energy consumption under typical operating conditions by implementing quantitative pre-cooling during off-peak electricity price periods based on ambient temperature prediction, and significantly reducing or stopping active cooling during peak periods by utilizing temperature decay. It directly couples the target pre-cooling temperature range with the battery cycle life, avoiding the surge in internal resistance caused by low temperatures and the material dissolution caused by high temperatures, thus maintaining a low capacity decay rate. This results in reduced auxiliary power losses and improved economic efficiency.

[0025] In another embodiment of this disclosure, step S2, training the prediction model, includes: using historical multi-source data as training samples; inputting the training samples into the prediction model and iterating the prediction model using a regression loss function to obtain the performance index of the current model training; when the number of iterations in which the performance index has not improved for a continuous period reaches a preset value, saving the model parameters of the current model to obtain the trained prediction model.

[0026] In this embodiment, firstly, multi-source data (including weather forecasts, historical ambient temperature data, and temperature conduction delay data of energy storage containers) are aggregated from historical databases. The data undergoes time alignment, missing value imputation, outlier removal, and standardization. Secondly, training, validation, and test sets are divided according to time series, and a backpropagation (BP) neural network is selected, using a regression loss function as the training objective. Thirdly, performance metrics are monitored on the validation set. When performance metrics do not significantly improve within a preset number of iterations k (e.g., k=5), the current model parameters are saved to obtain the trained prediction model. Finally, the trained model is deployed to the edge or cloud for online prediction, and is periodically (e.g., weekly or monthly) retrained or fine-tuned based on new data, with model versions recorded. The system's offline training and early-stop saving mechanism based on the validation set effectively prevent overfitting and achieves robust ambient temperature prediction capabilities. High-quality regression prediction provides a reliable basis for optimizing subsequent pre-cooling plans, thereby reducing ineffective cooling or safety risks caused by misjudgments and improving the economy and operational stability of off-peak scheduling.

[0027] In another embodiment of this disclosure, before step S4, which involves formulating a pre-cooling operation plan based on the pre-cooling target temperature range and the predicted ambient temperature, and adjusting the operating power according to the pre-cooling operation plan to reduce battery energy consumption, the method further includes: plotting a temperature decay curve based on the current battery characteristics; obtaining the battery pre-cooling time based on the temperature decay curve; and formulating a pre-cooling operation plan based on the pre-cooling time, the pre-cooling target temperature range, and the predicted ambient temperature.

[0028] In this embodiment, calibration experiments are conducted on-site for different PACK (battery pack) types (liquid cooling, air cooling) and protection levels (e.g., IP67). Temperature decay data is obtained by measuring the natural temperature recovery curve of the box over time under multiple ambient temperatures and cooling power conditions. Next, the obtained data is fitted, and an exponential or linear approximation model is selected to characterize the temperature decay. Then, the fitted temperature decay model is stored in the controller for online retrieval, and the sustainable time (i.e., the time required for the temperature to recover to the target upper limit after pre-cooling) is calculated by combining the current box temperature, the pre-cooling target temperature range, and the ambient temperature prediction value. Finally, the pre-cooling start time and depth are determined by using the sustaining time, electricity price period, and battery life constraints as inputs, and segmented pre-cooling, cold preservation, and supplementary cooling power curves are generated and executed. By quantifying the temperature decay curve, temperature inertia is transformed from an empirical property into an available engineering parameter. This enables pre-cooling decisions to predictably utilize natural recovery time to maximize peak-shaving benefits and minimize active cooling time, thereby improving energy efficiency and reducing the negative impact of frequent or excessive cooling on battery life. It also provides zoned parameter support for different PACK types.

[0029] Figure 2 This is an embodiment of the present disclosure. Figure 1 A schematic diagram of the method for adjusting operating power as described above; as follows: Figure 2 As shown, in another embodiment of this disclosure, step S4, which involves formulating a pre-cooling operation plan based on the pre-cooling target temperature range and the predicted ambient temperature, and adjusting the operating power according to the pre-cooling operation plan to reduce battery energy consumption, includes: step S41, obtaining electricity price information; step S42, adjusting the operating power to a preset first range during periods of low electricity prices; step S43, adjusting the operating power to a preset second range during periods of flat electricity prices; and step S44, adjusting the operating power to a preset third range during periods of high electricity prices.

[0030] In this embodiment, by periodically synchronizing with the electricity market or local electricity price interface to obtain electricity price curves for multiple future time periods, the time periods are roughly classified into off-peak, flat, and peak periods during engineering implementation, and the cooling power is constrained to the corresponding power ranges (e.g., 80% to 100% for off-peak, 30% to 50% for flat, and 10% to 20% for peak). Subsequently, the scheduling curve is issued for execution and corrected in real time during operation. By directly coupling the electricity price time series to power scheduling, the system can concentrate cooling or over-cool during low-price periods to reserve cooling capacity, thereby significantly reducing active cooling during high-price periods, thus achieving peak-valley arbitrage and significantly reducing operating electricity costs. By using an optimization solution method, feasible power curves that balance lifespan and economy can be found under multiple constraints, improving overall economic efficiency and scheduling precision.

[0031] In another embodiment of this disclosure, during periods of low electricity prices, the temperature is reduced to at least one degree below the lower limit of the pre-cooling target temperature range.

[0032] In this embodiment, when a current or upcoming off-peak electricity price window is identified, excess cooling is activated to reduce the temperature inside the box to at least 1°C below the lower limit of the pre-cooling target range during the pre-cooling phase. Controlled excess cooling can pre-store cooling capacity during low-price periods to replace active cooling during high-price periods, thereby significantly saving electricity costs and reducing peak energy consumption in typical scenarios.

[0033] In another embodiment of this disclosure, precooling is turned off when the predicted ambient temperature is less than a preset temperature.

[0034] In this embodiment, the predictive model outputs ambient temperature prediction vectors for multiple future time periods, and the controller compares these predictions with a pre-set preset temperature (e.g., 20°C). By introducing predictive judgment, unnecessary pre-cooling operations can be avoided when the environment already meets the temperature control target, thereby directly saving energy consumption and reducing wear and tear on equipment, improving the system's energy efficiency and operational economy.

[0035] In another embodiment of this disclosure, multi-source data includes weather forecasts, historical ambient temperature data for the same period, and temperature transmission delay data of the energy storage container.

[0036] In this embodiment, a multi-source data access layer is constructed and maintained. The main data sources include hourly forecasts from external meteorological services (ambient temperature, humidity, wind speed, shortwave radiation, etc.), historical ambient temperature datasets for the same period in the field (used for seasonality and weather type matching), historical temperature sequences inside the chamber and individual cells, and temperature conduction lag in the chamber (obtained through experimental calibration or online identification), real-time sensor data from the field (internal temperature, humidity, battery temperature, SOC, charging and discharging power), as well as electricity price and charging plan information. The data layer performs time-series alignment, interpolation, denoising, and feature extraction (including lag features, moving statistics, countdown to the start of charging, etc.), and the processed data is sent to the prediction module and the temperature decay modeling module. The fusion of multi-source data not only improves the accuracy of ambient temperature prediction but also clarifies the lag characteristics of ambient temperature changes being transmitted to the chamber through conduction delay data. This allows the pre-cooling plan to correctly select the lead time and excess cooling depth, ultimately significantly improving the reliability and energy saving effect of peak-shifting scheduling.

[0037] Figure 3 This is a schematic diagram of the energy storage and power consumption reduction control system described in the embodiments of this disclosure, as shown below. Figure 3 As shown, this disclosure also provides an energy storage and consumption reduction control system, including: Acquisition module 501 is used to acquire multi-source data and battery parameters; The prediction module 502 is used to input multi-source data into a pre-trained prediction model to make predictions and obtain the predicted ambient temperature value. The setting module 503 is used to set the pre-cooling target temperature range based on battery parameters to assess the battery life cycle. The adjustment module 504 is used to formulate a pre-cooling operation plan based on the pre-cooling target temperature range and the predicted ambient temperature, and adjust the operating power according to the pre-cooling operation plan to reduce energy consumption of battery storage.

[0038] This disclosure also provides a computer device, including a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the energy storage and power consumption reduction control method.

[0039] This disclosure also provides a computer-readable storage medium storing computer-readable instructions, which, when executed by a processor, implement the steps of an energy storage and power consumption reduction control method.

[0040] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0041] The above description is merely a specific embodiment of this disclosure, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not to be limited to the embodiments described herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for controlling energy storage consumption reduction, characterized in that, include: Acquire multi-source data and battery parameters; The multi-source data is input into a pre-trained prediction model to make predictions and obtain the predicted ambient temperature value. The target pre-cooling temperature range is set based on the battery parameters to assess the battery lifespan. A pre-cooling operation plan is formulated based on the pre-cooling target temperature range and the predicted ambient temperature, and the operating power is adjusted according to the pre-cooling operation plan to reduce energy consumption of battery storage.

2. The energy storage and consumption reduction control method according to claim 1, characterized in that, The training of the prediction model includes: Use historical multi-source data as training samples; The training samples are input into the prediction model, and the prediction model is iterated using a regression loss function to obtain the performance index of the current model training. When the number of iterations in which the performance metric has not improved reaches a preset value, the model parameters of the current model are saved to obtain the trained prediction model.

3. The energy storage and consumption reduction control method according to claim 1, characterized in that, Before the step of formulating a pre-cooling operation plan based on the pre-cooling target temperature range and the predicted ambient temperature, and adjusting the operating power according to the pre-cooling operation plan to reduce battery energy consumption, the method further includes: Plot the temperature decay curve based on the current battery characteristics; Based on the temperature decay curve, the battery pre-cooling time is obtained; The precooling operation plan is formulated based on the precooling time, the target precooling temperature range, and the predicted ambient temperature.

4. The energy storage and consumption reduction control method according to claim 1, characterized in that, The step of formulating a pre-cooling operation plan based on the pre-cooling target temperature range and the predicted ambient temperature, and adjusting the operating power according to the pre-cooling operation plan to reduce battery energy consumption, includes: Obtain electricity price information; During periods of low electricity prices, the operating power is adjusted to a preset first range; During the period of stable electricity prices, the operating power is adjusted to a preset second range; During peak electricity price periods, the operating power will be adjusted to a preset third range.

5. The energy storage and consumption reduction control method according to claim 4, characterized in that, During periods of low electricity prices, the temperature will be lowered to at least one degree below the lower limit of the pre-cooling target temperature range.

6. The energy storage and consumption reduction control method according to any one of claims 1 to 5, characterized in that, If the predicted ambient temperature is lower than the preset temperature, precooling is turned off.

7. The energy storage and consumption reduction control method according to any one of claims 1 to 5, characterized in that, The multi-source data includes weather forecasts, historical ambient temperature data for the same period, and temperature transmission delay data from energy storage containers.

8. An energy storage and consumption reduction control system, characterized in that, include: The acquisition module is used to acquire multi-source data and battery parameters; The prediction module is used to input the multi-source data into a pre-trained prediction model to make predictions and obtain the predicted ambient temperature value. The setting module is used to set the target pre-cooling temperature range based on battery parameters to assess the battery lifespan. The adjustment module is used to formulate a pre-cooling operation plan based on the pre-cooling target temperature range and the predicted ambient temperature, and adjust the operating power according to the pre-cooling operation plan to reduce energy consumption of battery storage.

9. A computer device, characterized in that, The device includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the energy storage and consumption reduction control method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the energy storage and consumption reduction control method as described in any one of claims 1 to 7.

Images