Display cabinet self-adaptive energy-saving control method based on distributed control

Abstract

The invention belongs to the technical field of adaptive energy-saving control, and particularly relates to a display cabinet adaptive energy-saving control method based on distributed control, and the method comprises the steps: building a communication module and a display cabinet, and obtaining the average energy consumption of the display cabinet in a region; building a data acquisition system inside and outside the display cabinet; designing a double-closed-loop energy-saving logic; display cabinet distributed control is built, and a dynamic adjustment task of local control node interaction data is obtained based on the real-time energy consumption of the display cabinet, the commodity heat-temperature coefficient and the current internal temperature of the display cabinet; acquiring equipment operation signal feature data in the data storage library, and establishing a fault prediction model; and based on the data on the cloud management node, predicting a use peak value by using the fault prediction model, and adjusting an operation strategy of the display cabinet. According to the invention, through combination of double-closed-loop energy-saving logic and regional average energy consumption calculation and energy-saving potential value sorting, accurate energy-saving regulation and control can be realized, energy configuration can be optimized, and operation cost can be reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of adaptive energy-saving control, and particularly relates to an adaptive energy-saving control method for display cabinets based on distributed control. Background Art

[0002] In modern commercial activities, display cabinets, as key devices for displaying various commodities, are widely used in places such as supermarkets, convenience stores, shopping malls, and various specialty stores. From cold drinks, dairy products, to baked goods, fresh flowers and green plants, to electronic products, jewelry, etc., different types of commodities rely on display cabinets to maintain a suitable storage environment and at the same time attract the attention of consumers. The widespread use of display cabinets has also brought significant energy consumption problems, and the electricity consumed by their continuous operation accounts for a relatively large proportion of the total energy consumption in commercial places.

[0003] In the current technological development trend, energy issues have attracted much attention. As high-energy-consuming devices in the commercial field, the development of energy-saving technologies for display cabinets is crucial. The rise of emerging technologies such as the Internet of Things has accelerated the intelligent management process of commercial equipment, enabling display cabinets to integrate into the intelligent system and always maintain a good operating state.

[0004] In the prior art, CN102176146B discloses an adaptive energy-saving control method for display cabinets, which belongs to the adaptive energy-saving control method. Its technical solution is: an adaptive energy-saving control device is installed on the display cabinet, and the device has a controller; a detection input circuit connected to the controller, a door switch sensor, an evaporator sensor, and a cabinet temperature sensor connected to the detection input circuit; a control output circuit connected to the controller, a compressor control circuit, a defrost control circuit, a lighting control circuit, a fan control circuit, and an alarm control circuit connected to the control output circuit.

[0005] However, the prior art, including the above patent, has problems such as poor control effect, inability to accurately predict equipment failures, and inability to effectively cope with usage peaks during the control process of display cabinets in actual applications. Summary of the Invention

[0006] In order to solve the above technical problems, the purpose of the present invention is to provide an adaptive energy-saving control method for display cabinets based on distributed control, which can realize precise energy-saving regulation, optimize energy allocation, and reduce operating costs by combining double-closed-loop energy-saving logic with regional average energy consumption calculation and energy-saving potential value ranking.

[0007] The present invention is achieved through the following technical solutions: An adaptive energy-saving control method for display cabinets based on distributed control includes the following steps: S1. Set up the communication module and the display cabinet. The setup of the display cabinet includes: establishing a collaborative framework for local control nodes, regional edge computing nodes, and cloud management nodes, and obtaining the average energy consumption of the display cabinets in the region based on the regional edge computing nodes; S2. Set up the data acquisition system inside and outside the display cabinet. The data obtained by the data acquisition system includes: the internal environmental data of the display cabinet, the external environmental data of the display cabinet, the equipment operation data, and the user behavior data; S3. Design a double-closed-loop energy-saving logic. The internal environment adjusts the compressor according to the intelligent control valve, and the external environment adjusts the local control parameters according to the deviation between the average energy consumption of the display cabinets in the region and the target; S4. Set up the distributed control of the display cabinet. Based on the real-time energy consumption of the display cabinet, the commodity heat temperature coefficient, and the current internal temperature of the display cabinet, obtain the dynamic adjustment task of the local control node interaction data; S5. Obtain the equipment operation signal characteristic data in the data repository and establish a fault prediction model; S6. Based on the data on the cloud management node, use the fault prediction model to predict the usage peak and adjust the operation strategy of the display cabinet.

[0008] Preferably, in S1, the local control node of the display cabinet real-time collects data such as temperature, humidity, light intensity, compressor running current, and the number of times the cabinet door is opened and closed, and sends the data to the regional edge computing node through the wireless communication module; the regional edge computing node collects the grid load rate data in the region and simultaneously receives the data uploaded by multiple local control nodes of the display cabinets; the cloud management node collects the comprehensive data uploaded by each regional edge computing node, including the average energy consumption of the display cabinets in the region, equipment fault warnings, external energy policies, and industry energy consumption standard data; between each node, the data collection frequency and strategy parameters are adjusted according to the actual energy-saving effect through the bidirectional communication feedback mechanism.

[0009] Preferably, the regional edge computing node receives the data uploaded by multiple local control nodes of the display cabinets and analyzes to obtain the average energy consumption of the display cabinets in the region , and the calculation formula is: ; where: n is the total number of display cabinets in the region, i is the number of the display cabinet in the region, is the self-energy consumption of the i-th display cabinet, is 's weight factor; The local control node transmits the collected data to the regional edge computing node in real time through the wireless communication module; at the same time, the local control node receives the control instructions sent by the regional edge computing node and adjusts the operation state of the display cabinet equipment according to the data instructions, adjusting the compressor frequency and turning on and off the lighting, thereby affecting the average energy consumption calculation of the regional edge computing node; The cloud management node collects the average energy consumption and device operation status data uploaded by the edge computing nodes in each region, and conducts data integration and analysis.

[0010] Preferably, in the step S2, a data acquisition system for the interior and exterior of the display cabinet is built, and the following data is obtained based on the local control node of the display cabinet: The interior environment data of the display cabinet includes: temperature, humidity, light intensity, the exterior environment data of the display cabinet includes: room temperature, humidity, light cycle, the device operation data includes: compressor running current, compressor running frequency, ventilation system running power, the user behavior data includes: user stay duration, cabinet door opening and closing times; Calculate the energy consumption threshold values inside and outside the display cabinet: ; Wherein, is the energy consumption threshold value, is 's weight factor, is the basic energy consumption threshold value, u is the parameter data number, is 's weight factor, is the u-th parameter data value, m is the total number of parameter data, and the local control node of the display cabinet feeds back the adjusted energy consumption data to the regional edge computing node.

[0011] Preferably, in the step S3, the inner loop control adjusts the compressor running frequency according to the intelligent control valve, which is expressed as: ; Wherein, k is the serial number of the control cycle, is the adjustment amount of the compressor running frequency in the k-th control cycle, is the proportionality coefficient, is the deviation value between the set temperature and the actual temperature in the k-th control cycle, is the integral coefficient, l is the summation index, is the accumulated sum of the temperature deviation in the control cycle, is the differential coefficient, is the deviation value between the set temperature and the actual temperature in the (k - 1)-th control cycle.

[0012] Preferably, the outer loop control area edge computing node adjusts the temperature set value and lighting brightness parameters of the local control node of the display cabinet according to the energy consumption deviation, and feeds back the inner loop control according to the data of the local control node of the display cabinet; The calculation formula for the energy consumption deviation of the area edge computing node is: ; Wherein, is the deviation value, is the average energy consumption of the display cabinets in the area, is the energy-saving target value.

[0013] Preferably, in the above-mentioned S4, based on the real-time energy consumption of the display cabinet, the commodity heat temperature coefficient, and the current temperature, the energy-saving task is allocated according to the following rules: S4.1. Analyze and obtain the energy-saving potential value of each display cabinet; The calculation formula for the energy-saving potential value of the display cabinet is: ; Wherein, is the energy-saving potential value of the i-th display cabinet, is the weight factor of , is the maximum allowable energy consumption value of the display cabinet, is the current actual energy consumption value of the i-th display cabinet, is the thermal stability coefficient of the commodities stored in the i-th display cabinet, is the weight factor of , is the temperature threshold preset for the display cabinet, is the current actual temperature of the display cabinet, is the weight factor of ; S4.2. Sort the display cabinets in descending order of the energy-saving potential value, and preferentially allocate energy-saving tasks to the display cabinets with large energy-saving potential values.

[0014] Preferably, in the above-mentioned S5, the fault prediction model takes the equipment operation signal as an important input. The current signal is processed, and the mean value, variance, peak value, and impulse factor characteristics are extracted. Based on the characteristics, the fluctuation situation and energy distribution of the current signal on the time scale are reflected, and the stability of the equipment operation is obtained. At the same time, by means of the frequency-domain Fourier transform analysis method, the current signal is transformed from the time domain to the frequency domain, and the main frequency, harmonic frequency, and corresponding amplitude characteristics are extracted. The frequency-domain characteristics help to reveal the vibration characteristics and potential mechanical fault signs during the equipment operation; The Fourier transform formula is: ; Wherein, t is the time variable, is the continuous signal in the frequency domain, is the representation form of the signal in the frequency domain, j is the imaginary unit, h is the angular frequency, is the complex exponential function.

[0015] Preferably, in the fault prediction model, for the parameter update of the fault prediction model, after the maintenance personnel complete the maintenance work on the display cabinet, they feedback the new equipment operation data, the information of the repaired and replaced parts, the adjusted internal environment data and external environment data of the display cabinet to the fault prediction model. The fault prediction model, based on the new data, uses the parameter update algorithm in machine learning to optimize and adjust based on the parameters inside the fault prediction model, so that the model can continuously and accurately predict the future faults of the display cabinet and adapt to the state changes after the display cabinet is maintained; The formula for the parameter update algorithm is: ; Wherein, represents the value of the parameter to be optimized after the (z + 1)-th iteration, is the value of the parameter at the z-th iteration, is the learning rate, is the objective function with respect to the parameter gradient.

[0016] Preferably, in S6, the cloud management node continuously collects and stores the energy consumption data, operation parameters, environmental parameters and historical usage peak data of each display cabinet in the area, uses the time series analysis algorithm, combines machine learning technology, and predicts the usage peak in a specific future time period. When it is predicted that a usage peak is about to occur, the cloud management node sends an instruction to the regional edge computing node. The regional edge computing node adjusts the equipment operation strategy of the display cabinet in advance according to the instruction and the actual situation of the local display cabinet. The specific adjustment methods include: reducing the refrigeration intensity of the display cabinet, adjusting the brightness and opening time of the lighting system, optimizing the start-stop logic of the compressor. At the same time, the regional edge computing node monitors the operation state of the display cabinet in real time and dynamically adjusts the operation strategy according to the feedback information.

[0017] The present invention has the following beneficial effects: By logically combining the regional average energy consumption calculation and the energy-saving potential value ranking through a double closed-loop link, the present invention can achieve precise energy-saving regulation, optimize energy allocation, and reduce operating costs; through the collaborative communication of local, regional and cloud nodes, as well as the processing of multi-source data, it can realize real-time monitoring, precise control and intelligent management of the operation state of the display cabinet; the fault prediction model uses equipment operation signal analysis to discover potential fault risks in advance, ensure the stable operation of the equipment, reduce downtime losses, and extend the service life; the cloud management node predicts the usage peak, enabling the display cabinet to better cope with peak hours, avoid excessive energy consumption and faults, and improve energy utilization efficiency and equipment reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is the method flow chart of the present invention; Figure 2 It is a flowchart for energy consumption control and data processing of a display cabinet in an embodiment of the present invention; Figure 3 It is a flowchart for fault prediction and peak usage response of a display cabinet in an embodiment of the present invention. Specific implementation manner

[0019] 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.

[0020] As Figure 1 shown, the adaptive energy-saving control method for a display cabinet based on distributed control includes the following steps: S1. Build a communication module and a display cabinet. The construction of the display cabinet includes: setting up a collaborative framework of local control nodes, regional edge computing nodes, and cloud management nodes, and obtaining the average energy consumption of the display cabinets in the region based on the regional edge computing nodes.

[0021] The local control node of the display cabinet collects data on temperature, humidity, light intensity, compressor operating current, and the number of times the cabinet door is opened and closed in real time, and sends it to the regional edge computing node through a wireless communication module.

[0022] The regional edge computing node collects data such as the grid load rate in the region, and at the same time receives data uploaded by multiple local control nodes of the display cabinet.

[0023] The cloud management node collects the comprehensive data uploaded by each regional edge computing node, including the average energy consumption of the display cabinets in the region, equipment fault warnings, external energy policy information, and industry energy consumption standard data; between each node, the data collection frequency and strategy parameters are adjusted according to the actual energy-saving effect through a two-way communication feedback mechanism.

[0024] The regional edge computing node receives the data uploaded by multiple local control nodes of the display cabinet, and analyzes to obtain the average energy consumption of the display cabinets in the region , and the calculation formula is: ; Where: n is the total number of display cabinets in the region, i is the number of the display cabinet in the region, is the self-energy consumption of the i-th display cabinet, is the weight factor of.

[0025] In the calculation of each formula in this embodiment, the dimension can be removed to simplify the calculation.

[0026] The local control node transmits the collected data to the regional edge computing node in real time through the wireless communication module; at the same time, the local control node receives the control instructions issued by the regional edge computing node and adjusts the operating state of the display cabinet equipment according to the data instructions, adjusting the compressor frequency and turning on / off the lighting, thereby affecting the average energy consumption calculation of the regional edge computing node.

[0027] The cloud management node collects the average energy consumption and equipment operating state data uploaded by each regional edge computing node for data integration and analysis.

[0028] The above settings are obtained from the database and are calculated based on historical data. and , establish and its corresponding mapping set of weight factors to obtain the current . The , , , , in the following text are also obtained through the mapping set of historical data and weight factors established in the database, that is, the corresponding weight factors are obtained according to the current data.

[0029] In this embodiment, during the construction of the display cabinet system, the local control node collects temperature, humidity and other data in real time and transmits them to the regional edge computing node. The regional edge computing node collects the regional power grid load rate data while receiving the data of multiple local control nodes, analyzes and calculates the average energy consumption of the display cabinets in the region, and issues instructions to the local control node to adjust the equipment operating state to affect the energy consumption. The cloud management node collects the comprehensive data such as the average energy consumption and equipment operating state uploaded by the regional edge computing node for integration and analysis. Through the two-way communication feedback mechanism between nodes, the data collection frequency and strategy parameters are adjusted according to the actual energy-saving effect.

[0030] Utilizing the distributed control system coordinated by the local control node, regional edge computing node, and cloud management node of the display cabinet is beneficial for the local control node to collect the internal data of the display cabinet. The regional edge computing node collects internal and external data and calculates the average energy consumption. There is two-way communication between nodes, and the data collection and strategy are adjusted according to the energy-saving effect. The cloud management node integrates and analyzes the data to provide support for decision-making. This system comprehensively collects data, accurately calculates energy consumption, and realizes dynamic optimization and adjustment, which helps the intelligent management and energy-saving and efficiency improvement of the display cabinet.

[0031] S2. Build a data collection system for the inside and outside of the display cabinet. The data obtained by the data collection system includes: the internal environment data of the display cabinet, the external environment data of the display cabinet, the equipment operation data, and the user behavior data.

[0032] In S2, a data acquisition system for the inside and outside of the display cabinet is built, and the following data is obtained based on the local control node of the display cabinet: The internal environmental data of the display cabinet includes: temperature, humidity, light intensity. The external environmental data of the display cabinet includes: room temperature, humidity, light cycle. The equipment operation data includes: compressor running current, compressor running frequency, ventilation system running power. The user behavior data includes: user stay duration, number of times the cabinet door is opened and closed; Calculate the energy consumption threshold values inside and outside the display cabinet: ; Among them, is the energy consumption threshold value, is 's weight factor, is the basic energy consumption threshold value, u is the number of the parameter data (each data obtained by the data acquisition system), is 's weight factor, is the value of the u-th parameter data, m is the total number of parameter data, and the local control node of the display cabinet feeds back the adjusted energy consumption data to the regional edge computing node.

[0033] The energy consumption threshold sets a reasonable range for the energy consumption of the display cabinet. When the actual energy consumption of the display cabinet exceeds this threshold, it indicates that its energy consumption is too high and energy-saving measures need to be taken; on the contrary, if the energy consumption is lower than the threshold, it means that the current energy consumption control is good, or there may be a situation of insufficient equipment operation.

[0034] The flowchart of the display cabinet energy consumption control and data processing is as Figure 2 shown.

[0035] In this embodiment, first, the local control node collects data on the internal and external environments, equipment operation, user behavior, etc. of the display cabinet and transmits it to the regional edge computing node. Then, a model is constructed and trained to learn the relationship between each parameter factor and energy consumption. According to the formula, the energy consumption threshold is calculated, and the parameter value, weight factor, basic energy consumption threshold, etc. are determined. Finally, the local control node feeds back the adjusted energy consumption data to the regional edge computing node to optimize the control strategy and adjust the parameters.

[0036] Using a neural network to integrate the internal and external environments, equipment operation, and user behavior data of the display cabinet to calculate the energy consumption threshold is beneficial to comprehensively cover the factors affecting energy consumption, accurately calculate the threshold using the weight factor, and provide a scientific quantitative index for energy-saving control. At the same time, the data feedback between the local control node and the regional edge computing node forms a closed loop, enhancing the pertinence and flexibility of energy-saving control, and can automatically adapt to changes in operating conditions, enabling the display cabinet to maintain good energy-saving performance in a complex environment.

[0037] S3. Design a double-closed-loop energy-saving logic. The internal environment adjusts the compressor according to an intelligent control valve, and the external environment adjusts the local control parameters according to the deviation between the average energy consumption of the display cabinets in the area and the target.

[0038] The internal loop control adjusts the operating frequency of the compressor according to the intelligent control valve, which is expressed as: ; where k is the serial number of the control period, is the adjustment amount of the operating frequency of the compressor in the kth control period, is the proportional coefficient, is the deviation value between the set temperature and the actual temperature in the kth control period, is the integral coefficient, l is the summation index, is the cumulative sum of the temperature deviations in the control period, is the differential coefficient, is the deviation value between the set temperature and the actual temperature in the (k - 1)th control period.

[0039] The external loop control area edge computing node adjusts the temperature setting value and lighting brightness parameter of the display cabinet local control node according to the energy consumption deviation, and feeds back to the internal loop control according to the data of the display cabinet local control node.

[0040] The calculation formula for the energy consumption deviation of the area edge computing node is: ; where, is the deviation value, is the average energy consumption of the display cabinets in the area, is the energy-saving target value.

[0041] In this embodiment, for the implementation of the double-closed-loop energy-saving logic, first, the internal loop control obtains the temperature deviation data and determines the control parameters in each control period, substitutes them into the formula to calculate the adjustment amount of the compressor operating frequency, and adjusts the frequency. The external loop control calculates the deviation between the average energy consumption of the display cabinets in the area and the energy-saving target value, adjusts the temperature setting value and lighting brightness parameter of the local control node accordingly, and optimizes the internal loop control based on the feedback from the local control node. The internal and external loops cooperate to achieve energy saving for the display cabinet.

[0042] The internal loop, based on the temperature deviation inside the display cabinet, precisely adjusts the operating frequency of the compressor through the proportional, integral, and differential coefficients to optimize the refrigeration energy consumption. The external loop, according to the deviation between the average energy consumption of the display cabinets in the area and the energy-saving target, intelligently adjusts the local control parameters to achieve overall energy saving. The internal and external loops cooperate with each other, enhancing the system stability and adaptability, reducing manual intervention, improving the intelligent level of the energy-saving control of the display cabinet, and enhancing the energy utilization efficiency.

[0043] S4. Build the distributed control of the display cabinet, and obtain the dynamic adjustment task of the local control node interaction data based on the real-time energy consumption of the display cabinet, the commodity heat temperature coefficient, and the current internal temperature of the display cabinet.

[0044] Based on the real-time energy consumption of the display cabinet, the commodity heat temperature coefficient, and the current temperature, allocate the energy-saving tasks according to the following rules: S4.1. Analyze and obtain the energy-saving potential value of each display cabinet.

[0045] The calculation formula for the energy-saving potential value of the display cabinet is: ; Among them, is the energy-saving potential value of the i-th display cabinet, is 's weight factor, is the maximum allowable energy consumption value of the display cabinet, is the current actual energy consumption value of the i-th display cabinet, is the thermal stability coefficient of the commodities stored in the i-th display cabinet, is 's weight factor, is the preset temperature threshold of the display cabinet, is the current actual temperature of the display cabinet, is 's weight factor.

[0046] S4.2. Sort the display cabinets in descending order of the energy-saving potential value, and preferentially allocate energy-saving tasks to the display cabinets with large energy-saving potential values.

[0047] In this embodiment, the local control node collects data such as the real-time energy consumption and the current temperature of the display cabinet, and obtains the preset relevant parameters. Substitute the data and the weight factor into the formula to calculate the energy-saving potential value of each display cabinet, sort the display cabinets in descending order of the energy-saving potential value, and preferentially allocate energy-saving tasks to the display cabinets with large energy-saving potential values.

[0048] Combining the real-time energy consumption, the commodity heat temperature coefficient, and the current temperature to calculate the energy-saving potential value, sorting the display cabinets according to this and preferentially allocating to those with large potential not only can accurately optimize the energy configuration, improve the energy utilization efficiency, but also enhance the adaptability of the system to the changes in the actual situation, and facilitate the manager to optimize the resource management. Preferentially allocate energy-saving tasks to the display cabinets with large energy-saving potential values, realize the high-efficiency energy saving under the distributed control of the display cabinet, and ensure the commodity quality and display effect while saving energy, and improve the shopping experience of consumers.

[0049] S5. Obtain the device operation signal characteristic data in the data storage repository and establish a fault prediction model.

[0050] The fault prediction model takes the device operation signals as important inputs. The current signal is processed, and features such as mean, variance, peak value, and impulse factor are extracted. Based on these features, the fluctuation and energy distribution of the current signal on the time scale are reflected, and the stability of the device operation is obtained. At the same time, by means of Fourier transform analysis in the frequency domain, the current signal is transformed from the time domain to the frequency domain, and features such as main frequency, harmonic frequency, and corresponding amplitude are extracted. The frequency domain features help to reveal the vibration characteristics and potential mechanical fault signs during device operation.

[0051] The Fourier transform formula is: ; where t is the time variable, is the continuous signal in the frequency domain, is the representation form of the signal in the frequency domain, j is the imaginary unit, h is the angular frequency, is the complex exponential function.

[0052] In the fault prediction model, for the parameter update of the fault prediction model, after the maintenance personnel complete the maintenance work on the display cabinet, the new device operation data of the display cabinet, the information of the replaced parts, the adjusted internal environment data and external environment data of the display cabinet are fed back to the fault prediction model. The fault prediction model, based on the new data, uses the parameter update algorithm in machine learning to optimize and adjust based on the parameters inside the fault prediction model. The model can continuously and accurately predict the future faults of the display cabinet and adapt to the state changes after the display cabinet is maintained.

[0053] The parameter update algorithm formula is: ; where, is the value of the parameter to be optimized after the (z + 1)-th iteration, is the value of the parameter at the z-th iteration, is the learning rate, is the objective function with respect to the parameter gradient.

[0054] In this embodiment, the operation signals of the display cabinet device are collected. The time domain processing of the current signal is performed to extract features such as mean to reflect the operation stability, and the frequency domain analysis is performed through Fourier transform to extract features such as main frequency to reveal potential faults. A model is selected and trained with the extracted features as inputs to establish a fault prediction model. When the device is maintained, the maintenance personnel feed back new data, and the model adjusts the parameters according to the parameter update algorithm to adapt to the device state changes; through model evaluation and continuous improvement, the fault prediction accuracy is continuously improved to ensure the stable operation of the display cabinet device.

[0055] By performing time-domain and frequency-domain analysis on the device operating current signal, extracting various features, constructing a fault prediction model, and accurately capturing the pre-fault signals of the device. At the same time, combined with the data feedback after equipment maintenance, the parameter update algorithm is used to dynamically optimize the model parameters to adapt to the changes in the device state. This mechanism can predict faults in advance, reduce downtime losses, optimize maintenance strategies, extend the service life of the device, and has the characteristics of accuracy, dynamic adaptability and economy.

[0056] S6. Based on the data on the cloud management node, use the fault prediction model to predict the usage peak and adjust the operation strategy of the display cabinet.

[0057] The cloud management node continuously collects and stores the energy consumption data, operation parameters, environmental parameters and historical usage peak data of each display cabinet in the area, uses the time series analysis algorithm, combined with machine learning technology, to predict the usage peak in a specific future time period.

[0058] The flowchart of fault prediction and usage peak response is as Figure 3 shown.

[0059] When it is predicted that the usage peak is about to occur, the cloud management node sends an instruction to the regional edge computing node. The regional edge computing node adjusts the device operation strategy of the display cabinet in advance according to the instruction and the actual situation of the local display cabinet.

[0060] The specific adjustment methods include: moderately reducing the cooling intensity of some non-critical display cabinets, adjusting the brightness and opening time of the lighting system, optimizing the start-stop logic of the compressor. At the same time, the regional edge computing node monitors the operation status of the display cabinet in real time and dynamically adjusts the operation strategy according to the feedback information.

[0061] In this embodiment, first, the cloud management node continuously collects and stores data such as the energy consumption, operation, environment and historical peak of the display cabinet, and then uses the time series analysis algorithm and machine learning technology to predict the future usage peak. When it is predicted that the peak is approaching, the cloud sends an instruction to the regional edge computing node. The regional edge computing node combines the local actual situation to adjust the operation strategies such as the cooling intensity, lighting system, and compressor start-stop logic of the display cabinet, and monitors the operation status in real time, and dynamically optimizes according to the feedback to achieve the balance between energy saving and normal operation.

[0062] Using the data of the cloud management node to predict the usage peak and adjust the device operation strategy, accurately predicting the electricity peak through time series analysis and machine learning, guiding the regional edge computing node to optimize the parameters of equipment such as refrigeration, lighting, and compressor in advance, and realizing the off-peak allocation of energy; at the same time, dynamically monitoring the feedback to ensure stable operation, which not only reduces energy consumption and saves costs, but also realizes intelligent management driven by data, and can flexibly adapt to the needs of multiple scenarios, balancing energy saving and business operation.

Claims

1. An adaptive energy-saving control method for a display cabinet based on distributed control, characterized in that, It includes the following steps: S1. Set up the communication module and the display cabinet. The construction of the display cabinet includes: setting up the collaborative framework of the local control node, the regional edge computing node, and the cloud management node, and obtaining the average energy consumption of the display cabinets in the region based on the regional edge computing node; S2. Set up the data acquisition system inside and outside the display cabinet. The data obtained by the data acquisition system includes: the internal environment data of the display cabinet, the external environment data of the display cabinet, the equipment operation data, and the user behavior data; S3. Design the double-closed-loop energy-saving logic. The internal environment adjusts the compressor according to the intelligent control valve, and the external environment adjusts the local control parameters according to the deviation between the average energy consumption of the display cabinets in the region and the target; S4. Set up the distributed control of the display cabinet. Based on the real-time energy consumption of the display cabinet, the commodity heat temperature coefficient, and the current internal temperature of the display cabinet, obtain the dynamic adjustment task of the local control node interaction data; S5. Obtain the equipment operation signal feature data in the data repository and establish a fault prediction model; S6. Based on the data on the cloud management node, use the fault prediction model to predict the usage peak and adjust the operation strategy of the display cabinet.

2. The adaptive energy-saving control method for display cabinets based on distributed control according to claim 1, characterized in that, In S1, the local control node of the display cabinet real-time collects data such as temperature, humidity, light intensity, compressor running current, and the number of times the cabinet door is opened and closed, and sends it to the regional edge computing node through the wireless communication module; the regional edge computing node collects the grid load rate data in the region and simultaneously receives the data uploaded by multiple local control nodes of the display cabinets; The cloud management node collects the comprehensive data uploaded by each regional edge computing node, including the average energy consumption of the display cabinets in the region, equipment fault warnings, external energy policies, and industry energy consumption standard data; Each node adjusts the data acquisition frequency and strategy parameters through the bidirectional communication feedback mechanism according to the actual energy-saving effect.

3. The adaptive energy-saving control method for a display cabinet based on distributed control according to claim 2, wherein The regional edge computing node receives the data uploaded by multiple local control nodes of display cabinets, and analyzes to obtain the average energy consumption of the display cabinets in the region , and the calculation formula is: ； Where: n is the total number of display cabinets in the area, and i is the number of the display cabinet in the area. is the self - energy consumption of the i - th display cabinet. is the weight factor of The local control node transmits the collected data to the regional edge computing node in real time through the wireless communication module; at the same time, the local control node receives the control instructions issued by the regional edge computing node and adjusts the operation state of the display cabinet equipment according to the data instructions, adjusting the compressor frequency and turning on and off the lighting, thereby affecting the average energy consumption calculation of the regional edge computing node; The cloud management node collects the average energy consumption and equipment operation state data uploaded by each regional edge computing node and conducts data integration and analysis.

4. The adaptive energy-saving control method for a display cabinet based on distributed control according to claim 2, characterized in that In S2, set up the data acquisition system inside and outside the display cabinet, and obtain the following data based on the local control node of the display cabinet: The internal environment data of the display cabinet includes: temperature, humidity, light intensity, the external environment data of the display cabinet includes: room temperature, humidity, light cycle, the equipment operation data includes: compressor running current, compressor running frequency, ventilation system running power, and the user behavior data subset includes: user stay duration, the number of times the cabinet door is opened and closed; Calculate the energy consumption threshold inside and outside the display cabinet: ； Among them, is the energy consumption threshold value, is 's weight factor, is the basic energy consumption threshold value, u is the parameter data number, is 's weight factor, is the value of the u-th parameter data, m is the total number of parameter data, and the local control node of the display cabinet feeds the adjusted energy consumption data back to the regional edge computing node.

5. The adaptive energy-saving control method for a display cabinet based on distributed control according to claim 1, wherein, In S3, the inner loop control adjusts the compressor running frequency according to the intelligent control valve, expressed as: ； where k is the sequence number of the control period, is the adjustment amount of the compressor operating frequency in the k-th control period, is the proportionality coefficient, is the deviation value between the set temperature and the actual temperature in the k-th control period, is the integral coefficient, and l is the summation index, is the cumulative sum of the temperature deviation in the control period, is the differential coefficient, is the deviation value between the set temperature and the actual temperature in the (k - 1)-th control period.

6. The adaptive energy-saving control method for a display cabinet based on distributed control according to claim 5, characterized in that The outer loop control regional edge computing node adjusts the temperature setting value and lighting brightness parameter of the local control node of the display cabinet according to the energy consumption deviation, and feeds back the inner loop control according to the data of the local control node of the display cabinet; The calculation formula for the energy consumption deviation of the regional edge computing node is: ； Among them, is the deviation value, is the average energy consumption of the display cabinets in the area, is the energy-saving target value.

7. The adaptive energy-saving control method for a display cabinet based on distributed control according to claim 1, characterized in that In S4, based on the real-time energy consumption of the display cabinet, the commodity heat temperature coefficient, and the current temperature, the energy-saving task allocation is carried out according to the following rules: S4.

1. Analyze and obtain the energy-saving potential value of each display cabinet; The calculation formula for the energy-saving potential value of the display cabinet is: ； Among them, is the energy-saving potential value of the i-th display cabinet, is 's weight factor, is the maximum allowable energy consumption value of the display cabinet, is the current actual energy consumption value of the i-th display cabinet, is the thermal stability coefficient of the goods stored in the i-th display cabinet, is 's weight factor, is the pre-set temperature threshold of the display cabinet, is the current actual temperature of the display cabinet, is 's weight factor; S4.

2. Sort the display cabinets in descending order of the energy-saving potential value, and preferentially allocate energy-saving tasks to the display cabinets with a large energy-saving potential value.

8. The adaptive energy-saving control method for a display cabinet based on distributed control according to claim 1, characterized in that, In S5, the fault prediction model takes the device operation signal as an important input. The current signal is processed, and the mean, variance, peak value, and pulse factor features are extracted. Based on the features, the fluctuation situation and energy distribution of the current signal on the time scale are reflected, and the stability of the device operation is obtained. At the same time, with the help of the frequency-domain Fourier transform analysis method, the current signal is transformed from the time domain to the frequency domain, and the main frequency, harmonic frequency, and corresponding amplitude features are extracted. The frequency-domain features help to reveal the vibration characteristics and potential mechanical fault signs during the device operation; The Fourier transform formula is: ； where t is the time variable, is a continuous signal in the frequency domain, is the representation of the signal in the frequency domain, j is the imaginary unit, and h is the angular frequency, is the complex exponential function.

9. The adaptive energy-saving control method for display cabinets based on distributed control according to claim 8, characterized in that: In the fault prediction model, for the parameter update of the fault prediction model, after the maintenance personnel complete the maintenance work on the display cabinet, the new device operation data of the display cabinet, the information of the replaced parts, the adjusted internal environment data and external environment data of the display cabinet are fed back to the fault prediction model. The fault prediction model optimizes and adjusts based on the new data using the parameter update algorithm in machine learning and the parameters inside the fault prediction model. The model can continuously and accurately predict the future faults of the display cabinet and adapt to the state changes after the display cabinet is maintained; The parameter update algorithm formula is: ； where, represents the value of the parameter to be optimized after the $(z + 1)$-th iteration, is the value of the parameter at the $z$-th iteration, is the learning rate, is the objective function with respect to the parameter gradient.

10. The adaptive energy-saving control method for a display cabinet based on distributed control according to claim 1, wherein: In S6, the cloud management node continuously collects and stores the energy consumption data, operation parameters, environmental parameters, and historical usage peak data of each display cabinet in the area. Using the time series analysis algorithm and combining machine learning techniques, it predicts the usage peak in a specific future time period. When it is predicted that the usage peak is about to appear, the cloud management node sends an instruction to the regional edge computing node. The regional edge computing node adjusts the device operation strategy of the display cabinet in advance according to the instruction and the actual situation of the local display cabinet. The specific adjustment methods include: reducing the refrigeration intensity of the display cabinet, adjusting the brightness and on-off time of the lighting system, optimizing the start-stop logic of the compressor. At the same time, the regional edge computing node monitors the operation state of the display cabinet in real time and dynamically adjusts the operation strategy according to the feedback information.

Images