Distributed energy-saving control method and system for refrigeration equipment

Abstract

The invention provides a distributed energy-saving control method and system for refrigeration equipment, and the method comprises the steps: obtaining physical parameters in an environment, fusing the preprocessed data, and generating an environment thermal field model through the mapping of an environment thermal field; according to a time sequence input sequence of the environment thermal field model, obtaining corresponding temperature field time sequence data, performing correction processing on the temperature field time sequence data, generating corresponding future temperature distribution, and forming a predicted future thermal field model; temperature deviation calculation analysis is conducted on future temperature distribution, the cooling capacity demand and the global cooling capacity demand of unit volume air are obtained, and an optimal refrigeration power distribution scheme is designed by minimizing the total energy consumption and meeting the power distribution constraint; and executing the optimal refrigeration power distribution scheme, determining the overall load balance degree of the system according to the current load ratio of each device, performing global optimization scheduling according to the overall load balance degree, and generating a final refrigeration device power distribution scheme.

Description

Technical Field

[0001] The present invention belongs to the field of distributed energy-saving control, and particularly relates to a distributed energy-saving control method and system for refrigeration equipment. Background Art

[0002] Refrigeration systems are widely used in scenarios such as data centers, industrial cooling, shopping malls, office buildings, and cold chain logistics. Its core goal is to maintain a stable temperature environment with the lowest energy consumption. However, there are still many problems in existing refrigeration control technologies, resulting in low overall energy efficiency. First, most refrigeration systems rely on local temperature sensors for adjustment, and can only obtain temperature information at specific measurement points, unable to accurately perceive the temperature distribution of the entire space, resulting in overcooling or overheating in local areas, and thus causing energy waste. Second, the current mainstream refrigeration control strategy usually adopts a passive response mechanism, that is, the adjustment is only started when the measured temperature exceeds the set threshold. This method has a large lag, cannot predict the temperature change trend, easily leads to large temperature fluctuations, and further increases unnecessary energy consumption. Finally, in the scenario of multiple refrigeration devices operating in coordination, each device is usually independently controlled, lacking a global optimization strategy, which may cause some devices to operate at high load for a long time, while other devices have a low load, resulting in uneven energy consumption distribution and reducing the overall system efficiency.

[0003] Therefore, how to introduce a more accurate environmental perception mechanism in the refrigeration system, combine temperature trend prediction for advanced adjustment, and achieve global energy saving through multi-device collaborative optimization has become a key technical challenge for improving the energy efficiency of the refrigeration system. Summary of the Invention

[0004] The object of the present invention is to propose a distributed energy-saving control method and system for refrigeration equipment, effectively overcoming the defects of existing refrigeration control technologies, and realizing more accurate and efficient energy-saving refrigeration control.

[0005] To achieve the above object, in the first aspect of the present invention, a distributed energy-saving control method for refrigeration equipment is provided. The method includes the following steps:

[0006] Obtain physical parameters in the environment, preprocess all physical parameters, and then generate an environmental thermal field model by fusing the preprocessed data through environmental thermal field mapping;

[0007] According to the time series input sequence of the environmental thermal field model, obtain the corresponding temperature field time series data, correct the temperature field time series data, generate the corresponding future temperature distribution, and form a predicted future thermal field model;

[0008] Perform temperature deviation calculation and analysis on the future temperature distribution to obtain the cooling demand of unit volume air and the global cooling demand, and design an optimal cooling power distribution plan with the goal of minimizing the total energy consumption and meeting the power distribution constraints; where the total energy consumption is the total power of the system, and the total power needs to meet the global cooling demand.

[0009] Among them, the satisfaction of power distribution constraints includes equipment operation restrictions, temperature demand constraints, and load balancing constraints; the equipment operation restriction is the maximum power of a single refrigeration device; the temperature demand constraint is the future cooling demand; the load balancing constraint is the power difference constraint between devices, which is used to ensure that no individual device operates overloaded.

[0010] Execute the optimal cooling power distribution plan, determine the overall load balancing degree of the system according to the current load ratio of each device, and perform global optimization scheduling according to the overall load balancing degree to generate the final cooling device power distribution plan.

[0011] Preferably, the physical parameters include temperature, humidity, and air velocity; the preprocessing includes anomaly detection and correction operations.

[0012] Preferably, the data after fusion preprocessing generates an environmental thermal field model through environmental thermal field mapping, including:

[0013] Obtain the space to be measured.

[0014] The measurement space is divided into M×N grid cells, and the temperature at the center point of each grid is calculated by weighted calculation of the surrounding measurement points to generate the temperature value of the grid center point at time t; among them, the weighting coefficient of the weighted calculation is calculated by exponential decay.

[0015] Calculate the gradient of adjacent grid points. If the gradient change is too large, adjust the temperature value of the grid center point at time t to make the temperature change smooth, reduce noise interference, calculate the local variance of the measurement points, dynamically adjust the filtering parameters, and remove high-frequency noise to ensure the stability of the thermal field.

[0016] Combine the optimized temperature values of all grid points to form the final thermal field M t ={T(x,y,t)∣(x,y)∈Ω}, where Ω is the regional range of the measurement space, and T(x,y,t) is the temperature value of point (x,y) at time t.

[0017] Preferably, the time series input sequence extracts the main change patterns by adaptive dimensionality reduction mapping to generate a reduced-dimensional feature matrix.

[0018] Preferably, the correction process of the temperature field time series data generates the corresponding future temperature distribution and constitutes a predicted future thermal field model, including:

[0019] Obtain the current heat field M t , and perform smoothing processing using an adaptive local smoothing term to generate a prediction model Among them, the adaptive local smoothing term is calculated through local gradient constraints to ensure that the predicted temperature change trend conforms to the heat diffusion law;

[0020] Calculate the prediction confidence based on the uncertainty measure calculated from historical errors;

[0021] Adjust the prediction model according to the prediction confidence to generate an adjusted prediction model

[0022] Combine the predicted temperature values of all grid points to construct the final future temperature distribution Among them, is the predicted temperature value at the time point (x, y) at the future time t + T'; Ω is: the regional range of the measurement space.

[0023] Preferably, the adjusting the prediction model according to the prediction confidence includes:

[0024] Introduce confidence weighting between the predicted value and the current temperature field; among them, when the prediction result confidence is the highest, it most depends on the preliminary prediction value calculated based on the feature matrix after dimensionality reduction of the time series input; when the prediction confidence is the lowest, it most tends to maintain the current temperature field to avoid excessive prediction deviation.

[0025] Preferably, perform temperature deviation calculation and analysis on the future temperature distribution, obtain the cooling demand of unit volume of air and the global cooling demand, and design an optimal refrigeration power distribution scheme by minimizing the total energy consumption and satisfying the power distribution constraint, specifically:

[0026] Determine the temperature deviation between the predicted future temperature value and the set target temperature, and the temperature deviation represents the temperature change amount that needs to be adjusted at the current position, and combine the air density, specific heat capacity and air flow velocity to calculate the cooling demand of unit volume of air and calculate the global cooling demand;

[0027] Obtain the total power of the system, which is provided by N refrigeration devices together, design the refrigeration power optimization target, and the goal is to minimize the total energy consumption while ensuring that the global cooling demand is met;

[0028] Design the power distribution constraint;

[0029] Generate an optimal refrigeration power distribution scheme.

[0030] Preferably, the generating the optimal refrigeration power distribution scheme includes at least one of the following:

[0031] A. Adopt a gradient adjustment strategy, initialize the power of each refrigeration device, calculate a preliminary allocation plan based on the global cooling demand, and gradually adjust the power output of each device to ensure the lowest overall energy consumption while meeting the temperature requirements. Calculate the rate of change of the objective function to judge the convergence situation. If the optimal conditions are met or the maximum number of iterations is reached, output the final power plan;

[0032] B. Combine the power output of each device with the global cooling demand to calculate the cooling capacity contribution ratio per unit power consumption of all devices, and preferentially allocate a larger proportion of the load to the devices with a higher cooling capacity contribution ratio per unit power consumption. Set a power limit for the devices with a lower cooling capacity contribution ratio per unit power consumption to avoid a decrease in the overall energy efficiency. During the allocation process, introduce a dynamic adjustment strategy so that the devices with lower energy efficiency only supplement the additional cooling demand when necessary;

[0033] C. If a device operates at a high load for a long time, appropriately adjust the power of adjacent devices to balance the temperature distribution, and set a dynamic power adjustment window to re-evaluate the device status every cycle to adapt to environmental changes and improve the long-term stability of the refrigeration system.

[0034] Preferably, when implementing the optimal refrigeration power allocation plan, determine the overall load balance degree of the system according to the current load ratio of each device, and perform global optimization scheduling according to the overall load balance degree to generate the final refrigeration device power allocation plan, which specifically includes:

[0035] Calculate the current load ratio of each device, that is, the ratio of the power output of the current device to its maximum power, and calculate the overall load balance degree of the system to measure the distribution of the loads of all devices;

[0036] Dynamically adjust the device power according to the distribution of the loads of all devices:

[0037] If the current load ratio of a device is much lower than the average load ratio of all devices, appropriately increase the power of this device to relieve the pressure on the high-load devices;

[0038] If the current load ratio of a device is much higher than the average load ratio of all devices, appropriately reduce the power of this device to prevent overloading and transfer the excess load to other devices;

[0039] If the temperature gradient in adjacent areas is large, increase the power of the devices close to the high-temperature area to accelerate temperature balance;

[0040] Adaptively optimize the operating state according to the distribution of the loads of all devices:

[0041] If a device is under low load for a long time, determine whether it can be shut down briefly to reduce unnecessary energy consumption; if a device is under high load for a long time, preferentially schedule other devices to increase power and reduce the operating pressure of this device;

[0042] Set a time window, recalculate the load balance degree every time window, and adjust the power distribution strategy according to the latest data;

[0043] Generate the final power distribution plan for the refrigeration equipment.

[0044] In another aspect of the present invention, there is provided a distributed energy-saving control system for refrigeration equipment, and the system includes:

[0045] An environmental thermal field model construction unit, which acquires physical parameters in the environment, preprocesses all physical parameters, and then generates an environmental thermal field model through environmental thermal field mapping by fusing the preprocessed data;

[0046] A future thermal field model construction unit, which is used to obtain corresponding temperature field time series data according to the time series input sequence of the environmental thermal field model, perform correction processing on the temperature field time series data, generate corresponding future temperature distributions, and constitute a predicted future thermal field model;

[0047] A refrigeration power distribution plan initialization unit, which is used to perform temperature deviation calculation and analysis on the future temperature distribution, obtain the cooling capacity demand per unit volume of air and the global cooling capacity demand, and design an optimal refrigeration power distribution plan with the goal of minimizing the total energy consumption and meeting the power distribution constraints; wherein, the total energy consumption is the total power of the system, and the total power needs to meet the global cooling capacity demand;

[0048] Among them, the power distribution constraints to be met include equipment operation restrictions, temperature demand constraints, and load balance constraints; the equipment operation restrictions are the maximum power of a single refrigeration equipment; the temperature demand constraints are the future cooling capacity demand; the load balance constraints are the power difference constraints between devices, which are used to ensure that no individual device operates overloaded;

[0049] A refrigeration power distribution plan optimization unit, which is used to execute the optimal refrigeration power distribution plan, determine the overall load balance degree of the system according to the current load ratio of each device, perform global optimization scheduling according to the overall load balance degree, and generate the final power distribution plan for the refrigeration equipment.

[0050] The beneficial technical effects of the present invention are at least as follows:

[0051] Aiming at the problems of limited sensing ability, control lag, and lack of collaborative optimization in the existing refrigeration system, the present invention proposes an adaptive energy-saving refrigeration control method and system based on environmental thermal field mapping.

[0052] First, by fusing various sensor data to construct an environmental thermal field map, the precise perception of the temperature distribution throughout the space is achieved, enabling the refrigeration system to optimize and control based on the global thermal information and avoid overcooling or overheating phenomena in local areas.

[0053] Secondly, temperature trend prediction technology is adopted to analyze historical temperature change data and infer the future temperature evolution trend, enabling the system to actively adjust the refrigeration power before the temperature fluctuates, reducing response lag and improving energy-saving effects.

[0054] Finally, in the scenario where multiple refrigeration devices operate together, a collaborative optimization mechanism is established. According to the environmental thermal field data and temperature prediction results, the refrigeration power of each device is dynamically adjusted to achieve load balancing, reduce overall energy consumption, and improve the stability and energy efficiency of the system operation. Through the above innovative points, the present invention can effectively overcome the defects of existing refrigeration control technologies and achieve more precise and efficient energy-saving refrigeration control. Description of the Drawings

[0055] The present invention is further described with reference to the drawings, but the embodiments in the drawings do not constitute any limitation to the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to the following drawings without creative efforts.

[0056] Figure 1 It is a flowchart of a distributed energy-saving control method for refrigeration equipment disclosed in an embodiment of the present invention. Detailed Embodiments

[0057] The embodiments of the present invention are described in detail below. The examples of the embodiments are shown in the drawings, where the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions throughout. The embodiments described below with reference to the drawings are exemplary and are only used to explain the present invention and should not be construed as a limitation to the present invention.

[0058] Embodiment 1

[0059] As Figure 1 shown, a distributed energy-saving control method for refrigeration equipment provided by an embodiment of the present invention includes the following steps:

[0060] S1. Obtain the physical parameters in the environment, preprocess all the physical parameters, and then fuse the preprocessed data through an environmental thermal field map to generate an environmental thermal field model.

[0061] Specifically, the goal of this step is to construct an environmental thermal field model M t, providing an accurate data basis for subsequent temperature prediction, power optimization, and coordinated scheduling of refrigeration equipment. Traditional refrigeration systems rely only on a few fixed sensors, making it difficult to comprehensively sense the ambient temperature, resulting in over-cooling or under-cooling in local areas and affecting the overall energy efficiency. To address this issue, this solution proposes a method of multi-source sensor fusion + spatial dynamic modeling + thermal field interpolation optimization to construct a high-precision spatial temperature field and ensure the spatio-temporal consistency of the data.

[0062] Furthermore, set measurement points:

[0063] Set N measurement points in the environment, and the measurement point set S = {s 1 , s 2 ,..., s N}.

[0064] Collect physical parameters such as temperature, humidity, and air velocity to form a data vector:

[0065]

[0066] Where:

[0067] The temperature value of measurement point s i at time t

[0068] The humidity value of measurement point s i at time t

[0069] The air velocity of measurement point s i at time t

[0070] Furthermore, anomaly detection and correction:

[0071] Calculate the temperature difference between measurement point s i and its neighboring measurement point s j . If it exceeds the set threshold, it is considered an anomaly:

[0072]

[0073] Where:

[0074] The set of neighboring measurement points of measurement point s i

[0075] The degree of anomaly of measurement point s i

[0076] For abnormal data points, joint correction using time interpolation and spatial interpolation is adopted:

[0077]

[0077] ​

[0078] Among them:

[0079] λ: Time regression factor (controls the weight of time interpolation)

[0080] w i,j : Measurement point s i and the neighboring measurement point s j The weighted coefficient between them

[0081] Furthermore, the construction of the thermal field:

[0082] Grid area division:

[0083] The measurement space is divided into M×N grid cells, and the temperature of the center point g of each grid k is calculated by weighted averaging of the surrounding measurement points:

[0084]

[0085] Among them:

[0086] The temperature value of the grid center point g k at time t

[0087] W i,k : The influence weight of the measurement point s i on the grid point g k i,k The weighted coefficient W

[0088] adopts exponential decay: i,k

[0089]

[0090] Among them:

[0091] d i,k : The Euclidean distance between the measurement point s i and the grid center g k

[0092]

[0093] α: Hyperparameter controlling the decay rate

[0094] Furthermore, the optimization of the thermal field:

[0095] Local gradient optimization:

[0095] Calculate the gradients of adjacent grid points. If the gradient changes too much, then adjust to make the temperature change smooth and reduce noise interference.

[0096] Adaptive spatio-temporal filtering:

[0097] Calculate the local variance of the measurement points, dynamically adjust the filtering parameters, remove high-frequency noise, and ensure the stability of the thermal field.

[0098] Furthermore, generate the final thermal field model:

[0099] Construct the thermal field model:

[0100] Combine the optimized temperature values of all grid points to form the final thermal field:

[0101] M t ={T(x,y,t)|(x,y)∈Ω} (6)

[0102] Where:

[0103] Ω: The regional range of the measurement space

[0104] T(x,y,t): The temperature value at point (x,y) at time t.

[0105] S2. According to the time series input sequence of the environmental thermal field model, obtain the corresponding temperature field time series data, perform correction processing on the temperature field time series data, generate the corresponding future temperature distribution, and form the predicted future thermal field model.

[0106] The goal of this step is to predict the temperature distribution at the future time t+T' based on the environmental thermal field model M t , providing accurate input for subsequent dynamic refrigeration power calculation. Traditional refrigeration systems only rely on current temperature feedback to adjust power, unable to predict the cooling demand in advance, resulting in lagged adjustment. This step constructs an adaptive temperature prediction mechanism through time series modeling + spatial correlation correction + prediction uncertainty compensation to improve the prediction accuracy and stability. Furthermore, select the thermal field data of the past T moments to form the time series input sequence:

[0107] X

[0108] X t ={M t-T ,M t-T+1 ,...,M t} (7)

[0109] Where:

[0110] X t : The temperature field time series data for prediction

[0111] M t-T to M t : The thermal fields of the past T moments

[0112] Dimensionality reduction optimization: Since the dimensionality of the thermal field data is high and the direct modeling calculation cost is large, the adaptive dimensionality reduction mapping Φ(·) is used to extract the main change patterns:

[0113] Z t = Φ(X t ) (8)

[0114] Where:

[0115] Z t : The feature matrix after dimensionality reduction

[0116] Φ(·): Dimensionality reduction transformation mapping, combining spatio-temporal correlation to ensure that key temperature patterns are not lost

[0117] Furthermore, spatio-temporal correlation correction:

[0118] Spatial correction term:

[0119] Traditional prediction methods usually assume that the temperature changes between measurement points are independent. However, in actual scenarios, the temperature changes are affected by factors such as air flow, equipment heat dissipation, and personnel activities, and there is strong spatial correlation.

[0120] The adaptive local smoothing term Ψ(M t ) is used, combined with the current thermal field M t , to optimize the prediction model:

[0121]

[0122] Where:

[0123] F(Z t ): The preliminary prediction value calculated based on the time series input Z t

[0124] Ψ(M t ): The spatial correction term calculated based on the current thermal field M t

[0125] Correction term calculation:

[0126] Ψ(M t ) is calculated through local gradient constraints to ensure that the predicted temperature change trend conforms to the heat diffusion law:

[0127]

[0128] Where:

[0129] The Laplace transform of the current thermal field, representing the diffusion trend of the temperature field

[0130] β: Balance coefficient, controlling the correction amplitude to avoid over-adjustment​​

[0131] Furthermore, the prediction confidence calculation:

[0132] Since the temperature change is affected by multiple factors and there is uncertainty in the predicted value, it is necessary to calculate the prediction confidence σ t , and dynamically adjust the prediction weight:

[0133]

[0134] Where:

[0135] ω t : Dynamic weight, ω t ∈[0,1]

[0136] σ t : Uncertainty measure calculated based on historical errors, σ t The larger it is, the higher the instability of the prediction result

[0137] Adaptive compensation mechanism:

[0138] Introduce confidence weighting between the predicted value and the current temperature field:

[0139]

[0140] When the prediction result has a high confidence (σ t is small), mainly rely on F(Z t );

[0141] When the prediction confidence is low (σ t is large), it is more inclined to maintain the current temperature field M t , to avoid excessive prediction deviation.

[0142] Furthermore, generate the final future thermal field model:

[0143] Combine the predicted temperature values of all grid points to construct the final future temperature distribution:

[0144]

[0145] Where:

[0146] Predicted temperature value at the future time point t+T′ at (x,y)

[0147] Ω: Region range of the measurement space

[0148] As the input of step 3, it is used to calculate the future optimal cooling power.

[0149] S3. Calculate the temperature deviation of the future temperature distribution to obtain the cooling demand of unit volume air and the global cooling demand, and design an optimal cooling power distribution scheme with the goal of minimizing the total energy consumption and meeting the power distribution constraints; where the total energy consumption is the total power of the system, and the total power needs to meet the global cooling demand.

[0150] The goal of this step is based on the future thermal field model predicted in step 2 to calculate the optimal cooling power P that each refrigeration device needs to distribute at the future moment t+T'. t This ensures the minimization of the overall energy consumption while meeting the temperature requirements. Traditional refrigeration systems usually adopt fixed power or simple on-off control, which are difficult to adapt to the dynamic changes of temperature requirements, easily lead to local overcooling or overheating, and increase energy waste. This solution ensures the accuracy and stability of the cooling power distribution through the cooling demand calculation + equipment energy efficiency optimization + load balancing mechanism.

[0151] Furthermore, calculate the future cooling demand:

[0152] Calculate the temperature deviation at each spatial position (x,y) at the future moment t+T':

[0153]

[0154] Where:

[0155] ΔT(x,y,t+T'): The temperature deviation, representing the temperature change amount that needs to be adjusted at the current position (x,y).

[0156] The future temperature value predicted in step 2.

[0157] T target : The set target temperature.

[0158] Calculate the cooling demand of unit volume air, considering the air density ρ, specific heat capacity c p and air flow velocity V(x,y):

[0159] Q(x,y,t+T') = ρc p V(x,y)ΔT(x,y,t+T') (15)

[0160] Where:

[0161] Q(x,y,t+T'): The cooling demand required at the current position (x,y) at the moment t+T' (unit: J).

[0162] ρ: Air density (unit: kg / m 3 ).

[0163] c p: Specific heat capacity of air (unit: J / kg·K).

[0164] V(x,y): Air flow velocity (unit: m / s).

[0165] Calculate the global cooling demand:

[0166]

[0167] where Ω is the entire region.

[0168] Furthermore, construct the optimization objective of refrigeration power:

[0169] Set the total power P of the system t , provided jointly by N refrigeration devices:

[0170]

[0171] where:

[0172] Refrigeration power of the i-th refrigeration device at time t (unit: W).

[0173] The goal is to minimize the total energy consumption while ensuring that the global cooling demand Q total (t + T′) is satisfied:

[0174]

[0175] where:

[0176] η i : Coefficient of Performance (COP) of the i-th refrigeration device.

[0177] Furthermore, optimize the power distribution constraint:

[0178] Device operation limit: The power output of each device is restricted by its own capacity to ensure that it does not exceed the maximum operating power of the device:

[0179]

[0180] where P max is the maximum power of a single refrigeration device.

[0181] Temperature demand constraint: The total refrigeration power must meet the future cooling demand:

[0182]

[0183] Load balancing constraint: The power output between different devices should be as balanced as possible:

[0184]

[0185] Among them, δ controls the power difference between devices to ensure that no individual device operates overloaded.

[0186] Furthermore, the iterative optimization method

[0187] Adopts a gradient adjustment strategy, initializes the power of each refrigeration device, and calculates a preliminary allocation plan based on the global cooling demand Q total (t + T′).

[0188] Gradually adjust While ensuring that the temperature demand is met, the overall energy consumption is minimized.

[0189] Calculate the rate of change of the objective function, judge the convergence situation. If the optimal conditions are met or the maximum number of iterations is reached, output the final power plan.

[0190] Furthermore, calculate the ratio of the cooling capacity contribution per unit power consumption of all devices, that is:

[0191]

[0192] Among them, γ i Reflects the energy efficiency performance of the device under the current load. The higher the value, the stronger the cooling capacity per unit power.

[0193] Give higher γ i Devices are allocated a larger proportion of the load, and a power upper limit is set for lower γ i Devices to avoid a decrease in the overall energy efficiency.

[0194] During the allocation process, introduce a dynamic adjustment strategy so that devices with lower energy efficiency only supplement additional cooling demand when necessary.

[0195] Furthermore, consider the load balance between devices. If a device operates at a high load for a long time, appropriately adjust the power of adjacent devices to balance the temperature distribution. Set a dynamic power adjustment window, re-evaluate the device status every cycle to adapt to environmental changes, and improve the long-term stability of the refrigeration system. After optimizing the refrigeration device power allocation plan P t As the input of step 4, ensure the stable operation of the system and optimize the energy consumption globally.

[0196] S4. Execute the optimal refrigeration power allocation plan, determine the overall load balance degree of the system according to the current load ratio of each device, and perform global optimization scheduling according to the overall load balance degree to generate the final refrigeration device power allocation plan.

[0197] Specifically, the goal of this step is based on the optimal refrigeration power allocation plan P calculated in step 3 t, perform intelligent scheduling among multiple refrigeration devices to enable the entire system to meet temperature requirements with the lowest energy consumption. Since the power of refrigeration devices is limited, directly implementing according to the optimal power plan may cause local device overload while other devices are idle, thus affecting refrigeration efficiency and system stability. Therefore, this solution ensures reasonable energy consumption distribution among devices and improves the efficiency and reliability of the overall refrigeration system through load balancing adjustment + device collaborative scheduling + dynamic power correction.

[0198] Furthermore, calculate the current load ratio of each device, that is, the current operating power of the device as a proportion of its maximum power P max :

[0199]

[0200] where:

[0201] The load ratio of the i-th device ranges from [0, 1], and the larger the value, the higher the load.

[0202] P max : The maximum operable power of a single device.

[0203] Calculate the overall load balance degree of the system to measure the load distribution of all devices:

[0204]

[0205] where:

[0206] is the average load ratio of the system to ensure uniform load among all devices.

[0207] γ is the energy efficiency adjustment coefficient, and the second term is used to penalize the power of low-energy-efficiency devices, prompting the system to preferentially allocate high-energy-efficiency devices.

[0208] The goal is to minimize Λ t , that is, while meeting the refrigeration demand, balance the device load and reduce the overall energy consumption.

[0209] Furthermore, increase the power of low-load devices: If the load ratio of a certain device is much lower than then appropriately increase the power of this device to relieve the pressure on high-load devices:

[0210]

[0211] where κ is the adjustment factor to ensure smooth power changes and avoid frequent start-stop of devices.

[0212] Limit the power of high-load devices: If the load ratio of a certain device Far higher than Then appropriately reduce the power of the device to prevent overload and transfer the excess load to other devices:

[0213]

[0214] Combined with temperature gradient correction: Considering the change in the ambient temperature around the device, if the temperature gradient between adjacent regions is large, increase the power of the device closer to the high-temperature region to accelerate temperature equilibrium:

[0215]

[0216] where β is the environmental response coefficient, indicating the rate of change of the temperature around the device.

[0217] Furthermore, intelligent control of device startup and shutdown:

[0218] If a certain device has a low load for a long time Judge whether the device can be shut down temporarily to reduce unnecessary energy consumption.

[0219] If a certain device has a high load for a long time Then give priority to scheduling other devices to increase power and reduce the operating pressure of this device.

[0220] Dynamic power adjustment window:

[0221] Set a time window ΔT, and recalculate the load balance degree Λ every ΔT t , and adjust the power distribution strategy according to the latest data.

[0222] This can adapt to the change of ambient temperature and keep the system always in the optimal operating state.

[0223] Furthermore, generate the final operation plan for the refrigeration device:

[0224] After the above optimization, generate the final power distribution plan for the refrigeration device:

[0225]

[0226] This plan ensures the load balance of all devices, avoids long-term overload of individual devices, and improves the system stability.

[0227] Through load balance adjustment, reduce unnecessary energy consumption and keep the best temperature control effect of the whole system at the lowest energy consumption.

[0228] This power distribution plan is finally applied to the device control system to execute the final refrigeration strategy.

[0229] Example 2

[0230] In another embodiment of the present invention, a distributed energy-saving control system for a refrigeration device is disclosed. The system includes:

[0231] An environmental thermal field model construction unit, which obtains physical parameters in the environment, preprocesses all physical parameters, and then generates an environmental thermal field model through environmental thermal field mapping by fusing the preprocessed data;

[0232] A future thermal field model construction unit, which is used to obtain corresponding temperature field time series data according to the time series input sequence of the environmental thermal field model, correct the temperature field time series data, generate a corresponding future temperature distribution, and form a predicted future thermal field model;

[0233] A refrigeration power distribution scheme initialization unit, which is used to calculate and analyze the temperature deviation of the future temperature distribution, obtain the cooling capacity demand per unit volume of air and the global cooling capacity demand, and design an optimal refrigeration power distribution scheme with the goal of minimizing the total energy consumption and meeting the power distribution constraints; wherein, the total energy consumption is the total power of the system, and the total power needs to meet the global cooling capacity demand;

[0234] Among them, the power distribution constraints include equipment operation restrictions, temperature demand constraints, and load balancing constraints; the equipment operation restrictions are the maximum power of a single refrigeration device; the temperature demand constraints are the future cooling capacity demand; the load balancing constraints are the power difference constraints between devices, which are used to ensure that no individual device operates overloaded;

[0235] A refrigeration power distribution scheme optimization unit, which is used to execute the optimal refrigeration power distribution scheme, determine the overall load balance degree of the system according to the current load ratio of each device, and perform global optimization scheduling according to the overall load balance degree to generate the final refrigeration device power distribution scheme.

[0236] The above describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be executed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily have to be in the specific order or continuous order shown to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0237] The systems, devices, modules, or units illustrated in the above embodiments can be specifically implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0238] For convenience of description, when describing the above devices, they are described separately as various units according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0239] Those skilled in the art should understand that the embodiments of this specification can be provided as a method, a system, or a computer program product. Therefore, the embodiments of this specification can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of this specification can take the form of a computer program product implemented 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.

[0240] This specification is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing devices to generate a machine, such that the instructions executed by the processors of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one or more flows and / or Figure 1 blocks or multiple blocks.

[0241] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device that implements the functions specified in Figure 1 one or more flows and / or Figure 1 blocks or multiple blocks.

[0242] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 steps of the functions specified in one block or multiple blocks.

[0243] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0244] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM) and / or non-volatile memory such as read-only memory (ROM) or flash memory (flash RAM). The memory is an example of computer-readable media.

[0245] Computer-readable media includes permanent and non-permanent, removable and non-removable media that can store information by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media such as modulated data signals and carrier waves.

[0246] It should also be noted that the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, commodity or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the statement "comprising one..." does not exclude the presence of additional identical elements in the process, method, commodity or device comprising the element.

[0247] This specification can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification can also be practiced in a distributed computing environment where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0248] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and reference can be made to the relevant parts of the method embodiments for the relevant content.

[0249] Finally, it should be noted that what is disclosed in an embodiment of a lithium battery pack chip equalization control platform of the present invention is only a preferred embodiment of the present invention, which is only used to illustrate the technical solution of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A distributed energy-saving control method for refrigeration equipment, characterized in that: The method comprises the following steps: Obtain the physical parameters in the environment, preprocess all the physical parameters, and then integrate the preprocessed data through the environmental thermal field mapping to generate the environmental thermal field model; According to the time series input sequence of the environmental thermal field model, corresponding temperature field time series data is obtained, the temperature field time series data is corrected, the corresponding future temperature distribution is generated, and a predicted future thermal field model is formed; Perform temperature deviation calculation and analysis on the future temperature distribution to obtain the cooling demand per unit volume of air and the global cooling demand, and design the optimal cooling power allocation scheme by minimizing the total energy consumption and satisfying the power allocation constraint; wherein the total energy consumption is the total power of the system, and the total power needs to meet the global cooling demand; The power allocation constraints are equipment operation restrictions, temperature demand restrictions and load balancing restrictions; the equipment operation restrictions are the maximum power of a single refrigeration equipment; the temperature demand restrictions are future cooling demand; the load balancing restrictions are power difference restrictions between equipment, which are used to ensure that no individual equipment is overloaded; Execute the optimal cooling power allocation plan, determine the overall load balance of the system according to the current load ratio of each device, perform global optimization scheduling based on the overall load balance, and generate the final cooling equipment power allocation plan.

2. A distributed energy-saving control method for refrigeration equipment according to claim 1, characterized in that: The physical parameters include temperature, humidity, and air flow rate; the preprocessing includes abnormality detection and correction operations.

3. A distributed energy-saving control method for refrigeration equipment according to claim 1, characterized in that: The fusion preprocessed data is mapped through the environmental thermal field to generate an environmental thermal field model, including: Get the space to be measured; The measurement space is divided into M×N grid units, and the temperature of each grid center point is weighted by the surrounding measurement points to generate the temperature value of the grid center point at time t; wherein the weighted coefficient of the weighted calculation is calculated using exponential decay; Calculate the gradient of adjacent grid points. If the gradient changes too much, adjust the temperature value of the grid center point at time t to make the temperature change smooth and reduce noise interference. Calculate the local variance of the measurement point, dynamically adjust the filter parameters, remove high-frequency noise, and ensure the stability of the thermal field. Combine the optimized temperature values ​​of all grid points to form the final thermal field M t ={T(x,y,t)|(x,y)∈Ω}, where Ω is the area range of the measurement space and T(x,y,t) is the temperature value of the point (x,y) at time t.

4. A distributed energy-saving control method for refrigeration equipment according to claim 1, characterized in that: The time series input sequence uses adaptive dimensionality reduction mapping to extract the main change mode and generate a feature matrix after dimensionality reduction.

5. A distributed energy-saving control method for refrigeration equipment according to claim 3, characterized in that: The correction processing of the temperature field time series data to generate the corresponding future temperature distribution and form a predicted future thermal field model includes: Get the current thermal field M t , using adaptive local smoothing terms for smoothing to generate a prediction model The adaptive local smoothing term is calculated by local gradient constraint to ensure that the predicted temperature change trend conforms to the law of thermal diffusion; Calculate forecast confidence based on uncertainty measures calculated from historical errors; The prediction model is evaluated based on the prediction confidence Make adjustments to generate an adjusted prediction model Combine the predicted temperature values ​​of all grid points to construct the final future temperature distribution in, is the predicted temperature value at the future time point (x, y) at time t+T′; Ω is the area range of the measurement space.

6. A distributed energy-saving control method for refrigeration equipment according to claim 4 or 5, characterized in that: The prediction model is evaluated according to the prediction confidence Make adjustments, including: Confidence weighting is introduced between the predicted value and the current temperature field. When the prediction result confidence is the highest, it is most dependent on the preliminary prediction value calculated based on the feature matrix after dimensionality reduction of the time series input. When the prediction confidence is the lowest, it is most inclined to maintain the current temperature field to avoid excessive prediction deviation.

7. A distributed energy-saving control method for refrigeration equipment according to claim 1, characterized in that: The temperature deviation calculation and analysis of the future temperature distribution is performed to obtain the cooling demand per unit volume of air and the global cooling demand, and the optimal cooling power allocation scheme is designed to minimize the total energy consumption and meet the power allocation constraints, specifically: Determine the temperature deviation between the predicted future temperature value and the set target temperature, the temperature deviation represents the temperature change amount that needs to be adjusted at the current position, and calculate the cooling demand per unit volume of air and the global cooling demand in combination with the air density, specific heat capacity and air flow velocity; Get the total power of the system, which is provided by N refrigeration equipments, and design the cooling power optimization target, which aims to minimize the total energy consumption while ensuring that the global cooling demand is met; Design power allocation constraints; Generate optimal cooling power allocation plan.

8. A distributed energy-saving control method for refrigeration equipment according to claim 1, characterized in that: Generating the optimal cooling power allocation scheme comprises at least one of the following: A. Adopt the gradient adjustment strategy, initialize the power of each refrigeration equipment, calculate the preliminary allocation plan based on the global cooling demand, and gradually adjust the power output of each device to ensure that the overall energy consumption is minimized while meeting the temperature requirements. Calculate the rate of change of the objective function and judge the convergence. If the optimal conditions are met or the maximum number of iterations is reached, output the final power plan; B. Calculate the unit power consumption and cooling capacity contribution ratio of all devices by combining the power output of each device with the global cooling capacity demand, and give priority to allocating a larger proportion of load to devices with high unit power consumption and cooling capacity contribution ratios, and set a power cap for devices with low unit power consumption and cooling capacity contribution ratios to avoid a decrease in overall energy efficiency. In the allocation process, introduce a dynamic adjustment strategy so that devices with lower energy efficiency can only supplement additional cooling capacity when necessary; C. If a device runs at high load for a long time, the power of adjacent devices should be adjusted appropriately to balance the temperature distribution, and a dynamic power adjustment window should be set to re-evaluate the device status in each cycle to adapt to environmental changes and improve the long-term stability of the refrigeration system.

9. A distributed energy-saving control method for refrigeration equipment according to claim 1, characterized in that: The optimal cooling power allocation scheme is executed, the overall load balance of the system is determined according to the current load ratio of each device, and global optimization scheduling is performed according to the overall load balance to generate the final cooling equipment power allocation scheme, which specifically includes: Calculate the current load ratio of each device, that is, the ratio of the current device's power output to its maximum power, and calculate the overall load balance of the system to measure the distribution of all device loads; Dynamically adjust device power based on the distribution of all device loads: If the current load ratio of a device is much lower than the average load ratio of all devices, the power of the device should be appropriately increased to reduce the pressure on high-load devices; If the current load ratio of a device is much higher than the average load ratio of all devices, the power of the device will be appropriately reduced to prevent overload, and the excess load will be transferred to other devices; If the temperature gradient between adjacent areas is large, increase the power of the equipment close to the high-temperature area to speed up temperature equalization; Adaptively optimize the operating status based on the distribution of all equipment loads: If a device has a low load for a long time, determine whether it can be shut down for a short time to reduce unnecessary energy loss; if a device has a high load for a long time, prioritize other devices to increase power and reduce the operating pressure of the device; Set a time window, recalculate the load balance in each time window, and adjust the power allocation strategy based on the latest data; Generate the final cooling equipment power allocation plan.

10. A distributed energy-saving control system for refrigeration equipment, characterized in that: The system comprises: The environment thermal field model building unit obtains the physical parameters in the environment, pre-processes all the physical parameters, and then integrates the pre-processed data through the environment thermal field mapping to generate the environment thermal field model; A future thermal field model construction unit is used to obtain corresponding temperature field time series data according to the time series input sequence of the environmental thermal field model, perform correction processing on the temperature field time series data, generate corresponding future temperature distribution, and form a predicted future thermal field model; A cooling power allocation scheme initialization unit is used to calculate and analyze the temperature deviation of the future temperature distribution, obtain the cooling demand per unit volume of air and the global cooling demand, and design the optimal cooling power allocation scheme by minimizing the total energy consumption and satisfying the power allocation constraint; wherein the total energy consumption is the total power of the system, and the total power needs to meet the global cooling demand; The power allocation constraints are equipment operation restrictions, temperature demand restrictions and load balancing restrictions; the equipment operation restrictions are the maximum power of a single refrigeration equipment; the temperature demand restrictions are future cooling demand; the load balancing restrictions are power difference restrictions between equipment, which are used to ensure that no individual equipment is overloaded; The cooling power allocation scheme optimization unit is used to execute the optimal cooling power allocation scheme, determine the overall load balance of the system according to the current load ratio of each device, perform global optimization scheduling according to the overall load balance, and generate the final cooling equipment power allocation scheme.

Images