Self-adaptive efficiency optimization air conditioner control system and method

Abstract

The invention discloses a self-adaptive efficiency optimization air conditioner control system and method, and relates to the technical field related to energy storage station air conditioner control. The air conditioner control system comprises a data acquisition module, a data analysis operation module, a hybrid prediction model control module and a power control module. The hybrid prediction model control module is used for establishing a hybrid prediction model based on the thermal characteristic dynamic analysis result and the thermal characteristic dynamic analysis parameters and predicting the temperature; the power control module is used for controlling the power of the air conditioner according to the temperature prediction result of the prediction model; according to the system and method, the temperature is accurately predicted by collecting and analyzing thermal characteristic data, building a hybrid model and dynamically switching a fusion model, and the prediction step length is dynamically adjusted in combination with a thermal time constant, so that the prediction step length is matched with the thermal inertia characteristic of the battery pack, the thermal inertia compensation is further optimized, and the service life of the battery is prolonged.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage station air-conditioning control, and particularly relates to an air-conditioning control system and method with adaptive efficiency optimization. Background Art

[0002] In an energy storage station, the battery is the core energy storage unit, and its performance and lifespan are greatly affected by the operating temperature. During the charging process of the battery, too high a temperature will trigger a series of adverse reactions. For example, too high a temperature will accelerate the chemical reaction rate inside the battery, resulting in an increased self-discharge phenomenon of the battery, thereby reducing the actual available capacity of the battery. At the same time, high temperature may also cause the electrolyte of the battery to decompose, generate gas, causing the internal pressure of the battery to rise, and even triggering safety hazards such as fire and explosion. On the other hand, when the temperature is too low, the internal resistance of the battery will increase significantly, which increases the energy loss during the charge and discharge process of the battery, greatly reduces the charging efficiency, and also shortens the cycle life of the battery.

[0003] Therefore, in order to ensure that the battery can operate efficiently, safely and with a long lifespan, it is crucial to accurately control the working temperature of the battery. This requires the air-conditioning control system to have extremely high control accuracy to maintain the battery operating in the optimal temperature range and meet the stringent requirements of the battery for temperature control accuracy.

[0004] Existing energy storage station air-conditioning control systems have exposed many key problems to be solved when dealing with complex working environments, especially low-temperature environments. Because existing energy storage station air-conditioning control systems mostly use linear control models to control the temperature, it is difficult to capture the non-linear temperature rise characteristics: Under low-temperature conditions, the electrochemical reactions inside the battery change, resulting in a significant increase in the internal resistance of the battery. The increase in the internal resistance of the battery makes the heat generation mechanism during the charge and discharge process of the battery more complex, showing obvious non-linear characteristics. Traditional temperature prediction models are often based on simple linear assumptions or fixed parameter models, and cannot accurately capture this non-linear trend of the temperature change of the battery pack caused by low temperature, resulting in a large deviation in the prediction of the battery temperature and making it difficult to meet the requirements of accurate temperature control in practical applications. Based on the above problems, the battery pack has a certain thermal inertia (the thermal inertia of the battery pack in the present invention refers to a lag characteristic exhibited by the battery pack during the heat transfer and temperature change processes), and its thermal time constant reflects the speed of temperature change of the battery pack. However, in existing air conditioning control systems, even though there are technologies that adopt non-linear prediction models, the prediction step lengths of such existing non-linear prediction models are often fixed and do not fully consider the dynamic changes of the thermal time constant of the battery pack. This results in a mismatch between the prediction step length and the actual thermal inertia characteristics of the battery pack. Since the thermal inertia characteristics of the battery pack are reflected by the thermal time constant, the thermal time constant describes the time required for the temperature of the battery pack to reach the final steady-state temperature after being subjected to a thermal disturbance. When the prediction step length is too long, the control strategy may lag behind the actual change of the battery temperature and cannot adjust the air conditioner in time to maintain the stability of the battery temperature; while when the prediction step length is too short, it may lead to over-adjustment of the control strategy, frequent start and stop of the air conditioner, which not only increases the energy consumption of the air conditioner but also may cause additional losses to the air conditioning equipment. The above problems are comprehensively reflected in the insufficient energy efficiency optimization of the air conditioning control system of the energy storage station, which does not have an adaptive ability and cannot flexibly adjust the operating power of the air conditioner according to the real-time state of the battery and environmental conditions. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides an air conditioning control system and method with adaptive efficiency optimization, which solves the control problems of the existing air conditioning control system of the energy storage station in dealing with complex working environments.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: On the one hand, the present invention discloses an air conditioning control system with adaptive efficiency optimization, including: A data acquisition module, used for dynamically acquiring multi-sensor data; A data analysis and operation module, used for performing dynamic thermal characteristic analysis and parameter calculation; A hybrid prediction model control module, used for establishing a hybrid prediction model based on the results of dynamic thermal characteristic analysis and dynamic thermal characteristic analysis parameters and predicting the temperature; A power control module, used for controlling the power of the air conditioner according to the result of temperature prediction by the prediction model; The hybrid prediction model control module includes: A neural network model sub-module, used for establishing and managing an LSTM neural network model; A physical heat conduction model sub-module, used for establishing and managing a physical heat conduction model; A prediction model control sub-module, used for controlling: When the working ambient temperature is greater than or equal to the first temperature threshold: control the LSTM model to predict the temperature, and correct the predicted value by the physical heat conduction model; When the working ambient temperature is less than the first temperature threshold: control the physical heat conduction model to predict the temperature, and the LSTM model provides boundary conditions; The prediction model control sub-module is also used to control the adaptive adjustment of the prediction step length in the temperature prediction of the LSTM neural network model or the physical heat conduction model.

[0007] On the other hand, the present invention discloses an air-conditioning control method for adaptive efficiency optimization, including the steps: Dynamically collect multi-sensor data; Conduct dynamic thermal characteristic analysis and parameter calculation; Based on the dynamic thermal characteristic analysis results and dynamic thermal characteristic analysis parameters, establish a hybrid prediction model and predict the temperature; Establish an LSTM neural network model; Establish a physical heat conduction model; When the working ambient temperature is greater than or equal to the first temperature threshold: control the LSTM model to predict the temperature, and correct the predicted value by the physical heat conduction model; When the working ambient temperature is less than the first temperature threshold: control the physical heat conduction model to predict the temperature, and the LSTM model provides boundary conditions; Adaptive adjustment of the prediction step length in the temperature prediction of the LSTM neural network model or the physical heat conduction model; Control the power of the air conditioner according to the result of the temperature prediction by the prediction model.

[0008] On the other hand, the present invention discloses an electronic device, including: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned air-conditioning control method for adaptive efficiency optimization.

[0009] The present invention provides an air - conditioner control system and method with adaptive efficiency optimization, having the following beneficial effects: By taking different time thresholds as intervals, the present invention calculates the thermal time constant of the battery pack at regular intervals to reflect the thermal inertia characteristics, calculates the internal resistance change rate in real - time to identify abnormal battery states, and calculates the heat transfer coefficient using the data of the heat - flow sensor to provide boundary conditions for the prediction model. Through this comprehensive and real - time acquisition and analysis of thermal characteristic data, it is possible to accurately grasp the thermal characteristic changes of the battery under different working conditions and establish a hybrid prediction model. This model can not only effectively capture the non - linear characteristics of battery temperature changes, but especially can predict the theoretical temperature change trend of the battery pack under low - temperature working conditions from the physical principle level. In a low - temperature environment, the physical heat conduction model can fully consider the heat transfer process between the battery pack and the environment. And importantly, by switching and fusing models, it can give full play to the advantages of the LSTM model in capturing non - linear characteristics, while skillfully using the advantages of the physical heat conduction model under low - temperature working conditions. Through this dynamic model - switching and fusion mechanism, it is possible to accurately predict the battery temperature under different working conditions. On the basis of the above effects, it is also possible to dynamically adjust the prediction step size in combination with the thermal time constant to match the thermal inertia characteristics of the battery pack, further optimizing the thermal inertia compensation and extending the battery life. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 It is a schematic diagram of the modules of the air - conditioner control system with adaptive efficiency optimization of the present invention; Figure 2 It is a schematic diagram of the flow of the air - conditioner control method with adaptive efficiency optimization of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0011] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0012] As Figure 1 , the present invention discloses an air - conditioner control system with adaptive efficiency optimization, which can be implemented in the form of hardware and / or software. The present invention preferably discloses a software application for an air - conditioner control system with adaptive efficiency optimization, which can be installed in electronic devices such as terminals or servers.

[0013] The present invention discloses an air - conditioner control system with adaptive efficiency optimization, including: A data acquisition module 100, used for dynamically acquiring multi - sensor data; A data analysis and operation module 101, used for performing dynamic thermal characteristic analysis and parameter calculation; The hybrid prediction model control module 102 is used to establish a hybrid prediction model based on the results of dynamic thermal characteristic analysis and dynamic thermal characteristic analysis parameters and predict the temperature; The power control module 103 is used to control the power of the air conditioner according to the result of temperature prediction by the prediction model; The hybrid prediction model control module includes: The neural network model sub-module 1020 is used to establish and manage the LSTM neural network model; The physical heat conduction model sub-module 1021 is used to establish and manage the physical heat conduction model; The prediction model control sub-module 1022 is used to control: When the working ambient temperature is greater than or equal to the first temperature threshold: control the temperature prediction by the LSTM model and correct the predicted value by the physical heat conduction model; When the working ambient temperature is less than the first temperature threshold: control the temperature prediction by the physical heat conduction model and provide boundary conditions by the LSTM model; The prediction model control sub-module 1022 is also used to control the adaptive adjustment of the prediction step length in the temperature prediction of the LSTM neural network model or the physical heat conduction model.

[0014] Optionally, the dynamic acquisition of multi-sensor data includes: Dynamically acquire temperature sensor data, humidity sensor data, battery internal resistance monitoring data, and heat flux sensor data; It also includes: synchronously acquiring the real-time power of the battery at fixed intervals and recording the operating parameters of the air conditioner at fixed intervals.

[0015] In implementation, the dynamic acquisition of multi-sensor data includes sensor array deployment: install high-precision temperature sensors (such as Pt100, accuracy ±0.3°C) at the top, middle, bottom, and low-temperature prone aggregation areas of the battery pack, and synchronously deploy environmental humidity sensors (accuracy ±0.5°C / ±3% RH), battery internal resistance monitoring modules (accuracy 0.1 mΩ), and heat flux sensors (accuracy ±5%).

[0016] During implementation, when the system starts a data acquisition task, it sequentially reads data from the temperature sensor, humidity sensor, battery internal resistance monitoring module, and heat flux sensor according to the set acquisition frequency. Before reading the data, the system first checks the status of the sensors to ensure that they are in normal working condition. For the data read from each sensor, preliminary format conversion and verification are performed to convert it into a data format that the system can recognize and process, and the integrity and validity of the data are checked. If data anomalies are found (such as missing data, data exceeding the normal range, etc.), the system automatically marks the data and records information such as the time when the anomaly occurred and the sensor number. The data after preliminary processing is stored in the system's database according to different sensor types and time sequences. The database adopts a distributed storage structure to improve the reliability and scalability of data storage. At the same time, to facilitate subsequent data query and analysis, the stored data is indexed and classified, and corresponding index tables and metadata information are established.

[0017] During implementation, the battery real-time power (current × voltage, accuracy 0.1% FS) is synchronously acquired at fixed intervals.

[0018] The air conditioner operation parameters (including compressor frequency, supply air temperature, and wind speed level) are recorded at fixed intervals.

[0019] Physical parameter initialization: Physically fixed parameters such as the heat capacity and heat dissipation area of the battery pack are recorded and stored.

[0020] During implementation, after the system starts, a fixed acquisition interval time is set, for example, 500 ms. When the set acquisition time point is reached, the system automatically sends a data request instruction to the battery current monitoring device and voltage monitoring device. After receiving the instruction, the current monitoring device and voltage monitoring device quickly read the battery current value and voltage value at the current moment. The read current value and voltage value are transmitted back to the system, and the battery real-time power is quickly calculated.

[0021] During implementation, a fixed recording interval is set, for example, a recording operation is performed every 2 seconds. The system sends a query instruction to the relevant data interface of the air conditioner control system to request information on the compressor frequency, supply air temperature, and wind speed level.

[0022] After receiving the instruction, the air conditioner control system reads in real time the current operating frequency data of the compressor stored in its internal memory. This frequency data is accurately measured by the frequency monitoring module in the compressor drive circuit and fed back to the control system. At the same time, it reads the supply air temperature data measured by the high-precision temperature sensor installed at the air outlet of the air conditioner, and the wind speed level data detected by the wind speed sensor. The accuracy of these sensors meets the system requirements.

[0023] Transmit the obtained compressor frequency, supply air temperature, and wind speed level data back to the main system. The system performs format conversion on these data to make them conform to the requirements of the internal data storage format of the system.

[0024] Optionally, perform dynamic thermal characteristic analysis and parameter calculation, including: Calculate the thermal time constant of the battery pack at intervals of the first time threshold to reflect the thermal inertia characteristic; Calculate the internal resistance change rate in real time, update the data at intervals of the second time threshold, and identify abnormal battery states; Calculate the heat transfer coefficient at intervals of the third time threshold using the data from the heat flux sensor, providing boundary conditions for the prediction model.

[0025] In implementation, perform dynamic thermal characteristic analysis and parameter calculation, including: Thermal time constant calculation: At intervals of the first time threshold (for example, if the first time threshold is 5 minutes, then at intervals of 5 minutes), calculate the thermal time constant of the battery pack regularly to reflect the thermal inertia characteristic.

[0026] The thermal time constant reflects the thermal inertia characteristic of the battery pack, which describes the time required for the battery pack to reach the final steady-state temperature after being thermally perturbed.

[0027] In the present invention, a first-order heat transfer model is adopted to calculate the thermal time constant. The battery pack can be regarded as a heat capacity body with a uniform temperature distribution, and its heat transfer process can be described by a first-order linear differential equation. According to the power balance equation at steady state, calculate the steady-state temperature, fit the collected temperature data with the theoretical formula, and the thermal time constant can be calculated and determined through the least squares method. The unit of the thermal time constant is a time unit.

[0028] Monitor the internal resistance change rate, calculate the internal resistance change rate in real time, update the data at intervals of the second time threshold (for example, if the second time threshold is 2 seconds, then update the data every 2 seconds), and identify abnormal battery states.

[0029] In implementation, obtain the real-time data of the battery internal resistance from the battery internal resistance monitoring module. The calculation of the internal resistance change rate is achieved by comparing the internal resistance measurement values at two adjacent time points.

[0030] Battery state abnormal identification: Preset a threshold for the internal resistance change rate to determine whether the battery state is abnormal. This threshold is determined according to the characteristics and historical data of the battery, and different types of batteries may have different thresholds. When the calculated internal resistance change rate exceeds the set threshold, the system will determine that the battery state is abnormal.

[0031] Determination of heat transfer coefficient: Using the data of the heat flux sensor, calculate the heat transfer coefficient at intervals of the third time threshold (for example, if the third time threshold is 10 minutes, calculate the heat transfer coefficient every 10 minutes), providing boundary conditions for the prediction model.

[0032] In implementation, read the heat flux data from the heat flux sensor. Heat transfer coefficient calculation: According to the law of heat conduction, the heat transfer coefficient h can be calculated by the formula h = q / (AΔT), where q is the heat flux density (which can be measured by the heat flux sensor), A is the heat dissipation area of the battery pack (recorded during the initialization of physical parameters), and ΔT is the temperature difference between the surface temperature of the battery pack and the ambient temperature. Through the measured heat flux data, the known heat dissipation area, and the real-time temperature difference, the heat transfer coefficient can be calculated.

[0033] Optionally, establishing a hybrid prediction model and temperature prediction includes: Establish an LSTM neural network model. By inputting the current battery temperature, power, environmental parameters, historical temperature change rate, real-time internal resistance, and thermal characteristic parameters into the LSTM neural network model, output the temperature prediction values for each time period within a number of future time periods (for example, within 30 minutes, 30 minutes is 6 time periods, and each time period is 5 minutes).

[0034] In implementation, establishing the LSTM neural network model includes: For the original temperature data collected from the temperature sensor, first perform denoising processing. Using the moving average filtering method, remove the high-frequency noise in the data to make the temperature data smoother. Then perform normalization processing, mapping the temperature data to the [0,1] interval for easy processing by the LSTM neural network model.

[0035] For the real-time power data calculated from the battery current and voltage, first perform accuracy verification to ensure that the power data is within a reasonable range. Then perform normalization processing in the same way, mapping it to the target numerical interval according to the maximum and minimum values of the power.

[0036] For environmental parameters such as environmental humidity, first perform validity checks to eliminate obvious errors or abnormal data. Then perform normalization operations according to their own value ranges to make them have a similar scale to other input data.

[0037] By recording the temperature data at different time points, calculate the temperature difference between adjacent time points, and then divide by the time interval to obtain the temperature change rate. To make the temperature change rate better reflect the overall trend, use the average temperature change rate within a certain time window as the input feature.

[0038] The mean squared error (MSE) is used as the loss function of the LSTM neural network model to measure the difference between the predicted temperature value and the actual temperature value of the model, and the parameters of the model are adjusted by minimizing the loss function.

[0039] The Adam optimization algorithm is used to update the weight parameters of the model. This algorithm can adaptively adjust the learning rate, accelerate the convergence speed of the model, and improve the training efficiency.

[0040] The hyperparameters of the LSTM neural network model are adjusted by the cross-validation method: the number of neurons in the hidden layer, the number of layers, and the time step.

[0041] Optionally, a hybrid prediction model is established and the temperature prediction includes: A physical heat conduction model is established. Based on the heat capacity of the battery pack, the heat dissipation area, the heat transfer coefficient, the ambient temperature, and the thermal time constant, a physical heat conduction model is established and used to predict the theoretical temperature change trend under low-temperature conditions.

[0042] The specific steps for establishing the physical heat conduction model include: The physical heat conduction model is established based on Fourier's law of heat conduction and the law of conservation of energy. Fourier's law describes the relationship between heat flux and temperature gradient, that is, the heat flux density is proportional to the temperature gradient; the law of conservation of energy ensures the balance of energy during the heat conduction process. For the battery pack as a thermal system, it can be regarded as a heat-conducting object with an internal heat source (heat generated by battery operation), and its heat conduction process can be described by a partial differential equation.

[0043] In the data acquisition stage, the heat capacity of the battery pack has been fixedly recorded and stored. This is a physical parameter reflecting the heat storage capacity of the battery pack, with the unit of joules per kelvin (J / K).

[0044] For the heat dissipation area (A): The heat dissipation area recorded during data acquisition is the surface area for heat exchange between the battery pack and the surrounding environment, with the unit of square meters.

[0045] For the heat transfer coefficient (h): The heat transfer coefficient calculated at intervals of the third time threshold (such as 10 minutes) using the heat flux sensor data, which represents the heat transfer ability between the battery pack and the surrounding environment, with the unit of watts per square meter kelvin (W / (m^2*K)).

[0046] For the ambient temperature: The temperature value of the surrounding environment collected in real time by the ambient temperature sensor, with the unit of degrees Celsius (℃).

[0047] The thermal time constant: The thermal time constant of the battery pack calculated at intervals of the first time threshold (such as 5 minutes), which reflects the thermal inertia characteristics of the battery pack, with the unit of minutes (min).

[0048] Then consider the heat conduction process of the battery pack and establish the following heat conduction equation: ; C is the heat capacity of the battery pack, T is the temperature of the battery pack, t is the time (s), h is the heat transfer coefficient, A is the heat dissipation area, T env is the ambient temperature, and P is the heat generation power (W) of the battery pack, which can be obtained from the real-time power data of the battery.

[0049] When starting the temperature prediction, take the actual temperature of the battery pack at the current moment as the initial temperature T 0 .

[0050] Boundary condition: Use the heat transfer coefficient h to describe the heat exchange between the battery pack and the surrounding environment, that is, the heat flux density q = h(T - T env ).

[0051] Adopt numerical solution methods, such as the Euler method or the Runge-Kutta method, to solve the heat conduction equation. The specific steps are as follows: Discretize the time, divide the prediction time period into several small time steps ∆t; Within each time step, according to the temperature T at the current moment n , use the heat conduction equation to calculate the temperature T at the next moment n + 1 : ; Repeat the above steps until the predicted time period is reached.

[0052] Under low-temperature working conditions, the internal resistance of the battery increases, and the heat generation power P will change. In order to accurately predict the theoretical temperature change trend under low-temperature working conditions, it is necessary to correct the heat generation power. According to the internal resistance monitoring data of the battery, calculate the internal resistance R of the battery in real time, and calculate the heat generation power according to Ohm's law P = I^2*R, where I is the battery current.

[0053] Optionally, controlling the temperature prediction by the LSTM model and correcting the predicted value by the physical heat conduction model includes: When the working ambient temperature is greater than or equal to the first temperature threshold (for example, the first temperature threshold is 5°C, and the working ambient temperature ≥ 5°C): Control the temperature prediction by the LSTM model and correct the predicted value by the physical heat conduction model (correction coefficient k = 0.9 - 1.1).

[0054] Specifically, correcting the predicted value by the physical heat conduction model means combining the temperature value predicted by the LSTM model with the theoretical temperature value calculated by the physical heat conduction model.

[0055] The specific calculation method is as follows: The corrected temperature prediction value = k × the prediction value of the LSTM model + (1 - k) × the theoretical temperature value of the physical heat conduction model. The corrected temperature prediction value will more accurately reflect the actual temperature change trend of the battery pack.

[0056] Optionally, controlling the temperature prediction by the physical heat conduction model and providing boundary conditions by the LSTM model includes: When the working ambient temperature is less than the first temperature threshold (for example, the first temperature threshold is 5°C, when the working ambient temperature < 5°C), control the temperature prediction by the physical heat conduction model and provide boundary conditions by the LSTM model.

[0057] The specific way of providing boundary conditions by the LSTM model is as follows: The temperature prediction values within several future time periods output by the LSTM model provide boundary references for the physical heat conduction model. For example, during the solution process of the physical heat conduction model, the temperature values predicted by the LSTM model are used as boundary constraints, making the prediction results of the physical heat conduction model more in line with the actual situation.

[0058] Optionally, the prediction model control sub-module 1022 is also used to control: When the temperature prediction error is greater than the second temperature threshold (for example, the second temperature threshold is 1.2°C) when the physical heat conduction model predicts the temperature and the LSTM model provides boundary conditions, then switch to the LSTM model for temperature prediction, and the physical heat conduction model corrects the prediction value.

[0059] Optionally, the adaptive adjustment of the prediction step length in the temperature prediction of the LSTM neural network model or the physical heat conduction model includes: First, initially set the prediction step length, and then adaptively adjust the prediction step length, that is, dynamically adjust the prediction step length according to the thermal time constant τ (for example, when the thermal time constant τ ≤ 15 minutes, the prediction step length is 20 minutes; when 15 minutes < τ ≤ 25 minutes, the prediction step length is 30 minutes; when τ > 25 minutes, the prediction step length is 40 minutes) to match the thermal inertia characteristics.

[0060] In implementation, the dynamic adjustment of the prediction step length according to the thermal time constant includes: After each calculation of the new thermal time constant, immediately update this value to ensure the accuracy of subsequent prediction step length adjustment.

[0061] The system has pre-set the prediction step lengths corresponding to different thermal time constant ranges, and when the thermal time constant meets the corresponding conditions, it is automatically adjusted to the corresponding prediction step length.

[0062] The adjusted prediction step will be applied to the subsequent temperature prediction of the LSTM neural network model or the physical heat conduction model, ensuring that the model can perform more accurate temperature prediction according to the thermal inertia characteristics of the battery pack.

[0063] Optionally, the power control module 103 is used to control the power of the air conditioner according to the result of the temperature prediction by the prediction model, including: When the result of the temperature prediction shows that there is a deviation between the predicted temperature and the ideal temperature (the ideal temperature refers to the optimal temperature for battery charging) within a certain future period, and the time with the deviation is greater than the fourth time threshold (for example, the fourth time threshold is 2 minutes); then control the power of the air conditioner to continuously operate at the maximum cooling / heating power; When the result of the temperature prediction shows that there is a deviation between the predicted temperature and the ideal temperature within a certain future period, and the time with the deviation is less than or equal to the fourth time threshold and greater than the fifth time threshold (for example, the fifth time threshold is 1 minute); then control the power of the air conditioner to continuously operate at the normal power (for example, the normal power is 80% of the maximum cooling / heating power); When the result of the temperature prediction shows that there is a deviation between the predicted temperature and the ideal temperature within a certain future period, and the time with the deviation is less than or equal to the fifth time threshold; then control the power of the air conditioner to continuously operate at the low-efficiency power (for example, the low-efficiency power is 50% of the maximum cooling / heating power).

[0064] In implementation, the process of controlling the power of the air conditioner involves the control of the air conditioner hardware. The query instructions sent by the air conditioner hardware control system read and transmit data such as the compressor frequency, supply air temperature, and wind speed level to the main system of the present invention, and at the same time receive the power control instructions sent by the main system of the present invention. The sent instructions reach the compressor, and the compressor adjusts the operating frequency to achieve different cooling / heating powers.

[0065] The various embodiments of the systems and technologies described above in this article can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: implemented in one or more computer programs, the one or more computer programs can be executed and / or interpreted on a programmable system including at least one programmable processor, the programmable processor can be a dedicated or general-purpose programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0066] AsFigure 2 , the present invention also discloses an air conditioner control method for adaptive efficiency optimization, and the method includes the steps: S100. Dynamically collect multi-sensor data; S101. Conduct dynamic thermal characteristic analysis and parameter calculation: S102. Establish a hybrid prediction model based on the dynamic thermal characteristic analysis results and dynamic thermal characteristic analysis parameters and predict the temperature: Establish an LSTM neural network model; Establish a physical heat conduction model; When the working ambient temperature is greater than or equal to the first temperature threshold: control the LSTM model to predict the temperature, and correct the predicted value by the physical heat conduction model; When the working ambient temperature is less than the first temperature threshold: control the physical heat conduction model to predict the temperature, and the LSTM model provides boundary conditions; Adaptive adjustment of the prediction step length in the temperature prediction of the LSTM neural network model or the physical heat conduction model; S103. Control the power of the air conditioner according to the result of the temperature prediction by the prediction model.

[0067] In some alternative embodiments, the dynamic thermal characteristic analysis and parameter calculation include: At intervals of the first time threshold, regularly calculate the battery pack thermal time constant to reflect the thermal inertia characteristic; calculate the internal resistance change rate in real time, update the data at intervals of the second time threshold, and identify abnormal battery states; use the thermal flow sensor data to calculate the heat transfer coefficient at intervals of the third time threshold to provide boundary conditions for the prediction model.

[0068] In some alternative embodiments, the LSTM neural network model is used to input the current battery temperature, power, environmental parameters, historical temperature change rate, real-time internal resistance and thermal characteristic parameters into the LSTM neural network model, and output the temperature prediction values for each time period within a plurality of future time periods.

[0069] In some alternative embodiments, establishing the physical heat conduction model includes: establishing a physical heat conduction model based on the battery pack heat capacity, heat dissipation area, heat transfer coefficient, environmental temperature, and thermal time constant and using it to predict the theoretical temperature change trend under low-temperature conditions.

[0070] In some alternative embodiments, when the temperature prediction error is greater than the second temperature threshold when the physical heat conduction model predicts the temperature and the LSTM model provides boundary conditions, then switch to the LSTM model to predict the temperature, and correct the predicted value by the physical heat conduction model.

[0071] In some alternative embodiments, the adaptive adjustment of the prediction step in the temperature prediction of the LSTM neural network model or the physical heat conduction model includes: first, initially setting the prediction step, and then adaptively adjusting the prediction step, that is, dynamically adjusting the prediction step according to the thermal time constant τ to match the thermal inertia characteristics.

[0072] In some alternative embodiments, controlling the power of the air conditioner according to the result of the temperature prediction by the prediction model includes: When the result of the temperature prediction shows that there is a deviation between the predicted temperature and the ideal temperature within a certain future period, and the time of the deviation is greater than the fourth time threshold; then control the power of the air conditioner to continuously operate at the maximum cooling / heating power; When the result of the temperature prediction shows that there is a deviation between the predicted temperature and the ideal temperature within a certain future period, and the time of the deviation is less than or equal to the fourth time threshold and greater than the fifth time threshold; then control the power of the air conditioner to continuously operate at the normal power; When the result of the temperature prediction shows that there is a deviation between the predicted temperature and the ideal temperature within a certain future period, and the time of the deviation is less than or equal to the fifth time threshold; then control the power of the air conditioner to continuously operate at the low-efficiency power.

[0073] The air conditioner control system with adaptive efficiency optimization provided by the embodiments of the present invention can execute the air conditioner control method with adaptive efficiency optimization provided by any embodiment of the present invention, and this method has the corresponding methods and beneficial effects of this system.

[0074] Obviously, the method of the present invention can be implemented through a computer program. The computer program for implementing the method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing devices, so that when the computer program is executed by the processor, the functions / operations specified in the flowchart and / or block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, executed partially on the machine and partially on a remote machine as an independent software package, or executed entirely on a remote machine or server.

[0075] Therefore, it can be understood that the present invention discloses an electronic device, including: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor can execute the above-mentioned air conditioner control method with adaptive efficiency optimization.

[0076] Finally, it should be noted that the above are only preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An air conditioning control system with adaptive performance optimization, characterized in that: include: Data acquisition module, used to dynamically collect multi-sensor data; Data analysis and calculation module, used for dynamic analysis of thermal characteristics and parameter calculation; A hybrid prediction model control module is used to establish a hybrid prediction model and predict the temperature based on the thermal characteristics dynamic analysis results and thermal characteristics dynamic analysis parameters; A power control module, used to control the power of the air conditioner according to the temperature prediction result of the prediction model; The hybrid prediction model control module includes: Neural network model submodule, used to build and manage LSTM neural network models; Physical heat conduction model submodule, used to establish and manage physical heat conduction models; Predictive model control submodule, used to control: When the working environment temperature is greater than or equal to the first temperature threshold: the temperature prediction by the LSTM model is controlled, and the prediction value is corrected by the physical heat conduction model; When the working environment temperature is lower than the first temperature threshold: the temperature prediction is controlled by the physical heat conduction model, and the boundary conditions are provided by the LSTM model; The prediction model control submodule is also used to control the adaptive adjustment of the prediction step size in the temperature prediction of the LSTM neural network model or the physical heat conduction model.

2. The air conditioning control system with adaptive performance optimization according to claim 1, characterized in that: The dynamic collection of multi-sensor data includes: Dynamically collect temperature sensor data, humidity sensor data, battery internal resistance monitoring data and heat flow sensor data; It also includes: synchronously collecting real-time battery power at fixed intervals, and recording air conditioner operating parameters at fixed intervals.

3. The air conditioning control system with adaptive performance optimization according to claim 1, characterized in that: The thermal characteristics dynamic analysis and parameter calculation include: At intervals of a first time threshold, a thermal time constant of the battery pack is calculated periodically to reflect thermal inertia characteristics; Calculate the internal resistance change rate in real time, update data at intervals of a second time threshold, and identify abnormal battery status; The heat exchange coefficient is calculated using the heat flow sensor data at intervals of the third time threshold to provide boundary conditions for the prediction model.

4. The air conditioning control system with adaptive performance optimization according to claim 1, characterized in that: The LSTM neural network model is used to output a temperature prediction value for each time period in the future by inputting the current battery temperature, power, environmental parameters, historical temperature change rate, real-time internal resistance and thermal characteristic parameters into the LSTM neural network model.

5. The air conditioning control system with adaptive performance optimization according to claim 1, characterized in that: The establishing and managing of the physical heat conduction model includes: establishing a physical heat conduction model based on the thermal capacity, heat dissipation area, heat exchange coefficient, ambient temperature, and thermal time constant of the battery pack and using it to predict the theoretical temperature change trend under low temperature conditions.

6. The air conditioning control system with adaptive performance optimization according to claim 1, characterized in that: The prediction model control submodule is also used to control: When the temperature is predicted by the physical heat conduction model and the boundary conditions are provided by the LSTM model, if the predicted temperature error is greater than the second temperature threshold, the temperature prediction by the LSTM model is switched, and the predicted value is corrected by the physical heat conduction model.

7. The air conditioning control system with adaptive performance optimization according to claim 1, characterized in that: The adaptive adjustment of the prediction step length in the temperature prediction of the LSTM neural network model or the physical heat conduction model includes: firstly setting the prediction step length, then adaptively adjusting the prediction step length, and dynamically adjusting the prediction step length according to the thermal time constant to match the thermal inertia characteristics.

8. The air conditioning control system with adaptive performance optimization according to claim 1, characterized in that: The controlling the power of the air conditioner according to the result of the temperature prediction by the prediction model comprises: When the result of temperature prediction shows that the predicted temperature within a certain period in the future deviates from the ideal temperature, and the time of the deviation is greater than the fourth time threshold, the power of the air conditioner is controlled to continuously operate at the maximum cooling / heating power; When the temperature prediction result shows that the predicted temperature within a certain period in the future deviates from the ideal temperature, and the time of the deviation is less than or equal to the fourth time threshold and greater than the fifth time threshold; the power of the air conditioner is controlled to be normal power and continue to run; When the temperature prediction result shows that the predicted temperature within a certain period in the future deviates from the ideal temperature, and the time of the deviation is less than or equal to the fifth time threshold, the power of the air conditioner is controlled to run continuously at low-efficiency power.

9. An air conditioning control method with adaptive performance optimization, characterized in that: Includes steps: Dynamically collect multi-sensor data; Conduct dynamic analysis of thermal characteristics and parameter calculation; Based on the dynamic analysis results and parameters of thermal characteristics, a hybrid prediction model is established to predict the temperature; Establish LSTM neural network model; Establish a physical heat conduction model; When the working environment temperature is greater than or equal to the first temperature threshold: the temperature prediction by the LSTM model is controlled, and the prediction value is corrected by the physical heat conduction model; When the working environment temperature is lower than the first temperature threshold: the temperature prediction is controlled by the physical heat conduction model, and the boundary conditions are provided by the LSTM model; Adaptive adjustment of prediction step size in temperature prediction of LSTM neural network model or physical heat conduction model; The power of the air conditioner is controlled according to the temperature prediction results of the prediction model.

10. An electronic device, comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, wherein the instructions are executed by the at least one processor to enable the at least one processor to perform the method of claim 9.

Images