Freezing and refrigerating Internet of Things intelligent temperature control and energy consumption management system

Abstract

The invention relates to the technical field of Internet of Things and refrigeration control, and discloses a freezing and refrigerating Internet of Things intelligent temperature control and energy consumption management system which comprises a plurality of distributed execution and sensing modules, an environment data acquisition module, a central processing and decision module and an energy and communication support module. The distributed execution and sensing module has dual modes of refrigeration and acoustic detection. The system obtains acoustic response data in a space through an acoustic detection mode of a distributed module, and a central processing and decision module inverts physical environment states such as cargo loading based on the data, fuses collected temperature and humidity data, and establishes an optimization model with minimum total energy consumption of the system as a target. And solving an optimal control instruction for each module, and issuing and executing the optimal control instruction. According to the invention, prospective environmental perception is carried out through the acoustic function of the multiplexing module, and the physical state is introduced into global energy consumption optimization as a feedforward parameter, so that the conversion from passive temperature control to active predictive energy consumption management is realized, and the operation energy consumption is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet of Things and automatic control, in particular to a frozen refrigeration Internet of Things intelligent temperature control and energy consumption management system. BACKGROUND

[0002] The temperature control system of existing frozen refrigeration facilities, such as cold storage or refrigerated vehicles, mostly relies on passive adjustment strategies based on temperature feedback. Such systems usually monitor the ambient temperature through a limited number of temperature sensors arranged in the space. When the monitored temperature deviates from the preset range, the control system starts or stops the centralized refrigeration compressor set to maintain the average temperature in the space.

[0003] This control method has its inherent limitations. First, its control logic is reactive, and the system cannot anticipate the root causes of temperature changes, such as the opening of the door, the introduction of a large amount of uncooled goods, etc. The system can only respond after the overall temperature has changed significantly. This lag not only leads to large fluctuations in the temperature of the storage space, affecting the storage quality, but also often triggers the long-time full-load operation of the refrigeration equipment to compensate for the temperature difference, causing unnecessary energy consumption.

[0004] In addition, traditional centralized refrigeration systems usually provide undifferentiated overall cooling, and lack the ability to fine-tune the local hot spots formed inside the space due to goods stacking, poor air flow, etc. To ensure that the hottest spot meets the temperature requirements, the manager often needs to set the temperature of the entire space lower, which further exacerbates energy waste. Although increasing the number of sensors can provide more detailed temperature distribution, it will increase the complexity and cost of wiring and maintenance. Therefore, how to realize the forward-looking perception of the physical state inside the frozen refrigeration space and fine-tune the control based on energy optimization is a technical problem that needs to be solved in this field. SUMMARY

[0005] In view of the shortcomings of the prior art, the present application provides a frozen refrigeration Internet of Things intelligent temperature control and energy consumption management system, which solves the problem that the existing frozen refrigeration temperature control technology mainly relies on lagging temperature feedback for passive adjustment, lacks forward-looking perception of the physical state inside the space and fine-tuning ability, and thus leads to high energy consumption and poor temperature uniformity.

[0006] To achieve the above purpose, the present application realizes the following technical scheme: a frozen refrigeration Internet of Things intelligent temperature control and energy consumption management system, comprising: a plurality of distributed execution and sensing modules, each of which is configured to provide a refrigeration capacity in a refrigeration mode and to emit and receive acoustic waves to obtain acoustic response data in an acoustic detection mode; a plurality of environmental data acquisition modules configured to acquire temperature and humidity data of a space in which the system is located; a central processing and decision module communicatively connected to the distributed execution and sensing modules and the environmental data acquisition modules, the central processing and decision module configured to receive the temperature and humidity data and the acoustic response data, to inverse the physical environment state in the space based on the acoustic response data, and to establish an optimization model aiming at minimizing the total energy consumption of the system in combination with the temperature and humidity data, and to determine control instructions for each of the distributed execution and sensing modules; an energy and communication support module configured to provide energy supply and communication support for the system.

[0007] Preferably, the distributed execution and sensing module is a thermo-acoustic refrigeration module, which specifically includes an acoustic driver configured to generate acoustic waves, a resonant tube configured to resonate acoustic waves, a stack configured to generate a temperature difference in an acoustic field, a cold end heat exchanger located at a low temperature side of the stack, and a sound pressure sensor configured to measure acoustic response in the acoustic detection mode.

[0008] Preferably, the central processing and decision module specifically includes: a data fusion and three-dimensional temperature field reconstruction unit configured to generate a three-dimensional temperature field model based on the temperature and humidity data; an acoustic state inversion unit configured to inverse the physical environment state based on the acoustic response data; a feedforward optimization decision unit configured to establish and solve the optimization model based on an evaluation result of the three-dimensional temperature field model and the physical environment state; an instruction generation and scheduling unit configured to convert a solution of the optimization model into the control instructions and to issue the control instructions.

[0009] Preferably, the acoustic state inversion unit specifically includes: obtaining an acoustic deviation vector by comparing a real-time room impulse response with a preset reference room impulse response, and inversely analyzing a cargo loading state or a gas composition anomaly in the space based on the acoustic deviation vector as an evaluation result of the physical environment state.

[0010] Preferably, the feedforward optimization decision unit specifically includes: fusing a reactive refrigeration demand calculated based on the three-dimensional temperature field model, a forward-looking refrigeration demand calculated based on an evaluation result of the physical environment state output by the acoustic state inversion unit, and generating a total refrigeration demand for the optimization model.

[0011] Preferably, the optimization model established by the feedforward optimization decision unit comprises the following constraint conditions: refrigeration demand constraints satisfying the overall refrigeration demand, and module capacity constraints of each distributed execution and perception module.

[0012] Preferably, the energy and communication support module provides a hybrid communication network architecture, which comprises a wired backbone network for connecting the central processing and decision module and the plurality of distributed execution and perception modules, and a wireless sensing sub-network for connecting the one or more environment data acquisition modules.

[0013] Preferably, the energy and communication support module further comprises a direct-current power bus for centrally converting external alternating current into direct current and supplying power to the plurality of distributed execution and perception modules and the central processing and decision module.

[0014] Preferably, the data fusion and three-dimensional temperature field reconstruction unit is configured to generate a three-dimensional temperature field model of the space by a spatial interpolation algorithm using the spatially dispersed temperature and humidity data.

[0015] A refrigeration and freezing Internet of Things intelligent temperature control and energy consumption management method comprises the following steps: S1, acquiring acoustic response data and temperature and humidity data in a space by a distributed execution and perception module and an environment data acquisition module, respectively; S2, inverting a physical environment state in the space based on the acoustic response data; S3, combining the physical environment state and the temperature and humidity data, establishing and solving an optimization model with the minimum system total energy consumption as the target, and generating control instructions for each distributed execution and perception module; S4, issuing the control instructions to drive the distributed execution and perception module to perform refrigeration operation.

[0016] The present application provides a refrigeration and freezing Internet of Things intelligent temperature control and energy consumption management system, which has the following beneficial effects: 1, the present application realizes the reduction of system energy consumption through the cooperative work of its acoustic state inversion unit and feedforward optimization decision unit; the system uses the acoustic detection mode of the distributed execution and perception module to non-contact perceive the newly added heat load in the warehouse, and inputs this information as a forward-looking refrigeration demand into the optimization model; so that the system control strategy can change from passive temperature compensation to active load prediction, and the refrigeration capacity is optimized and distributed in advance, avoiding excessive refrigeration or frequent start and stop of equipment due to temperature lag response, thereby reducing the total operation energy consumption of the system.

[0017] 2、The present application improves the uniformity and accuracy of space temperature control by using multiple distributed execution and perception modules, combined with data fusion and three-dimensional temperature field reconstruction unit; the system can generate a refined three-dimensional temperature field model, accurately locate the local hot spots in the space; based on this model, the central processing and decision module can issue accurate cooling instructions to the specific distributed module closest to the hot spot, realizing targeted adjustment of the temperature field; this method replaces the traditional centralized cooling system's large-scale, non-differentiated cold delivery, effectively eliminating temperature dead angles and ensuring the uniformity and stability of the entire storage space temperature field.

[0018] 3、The present application enhances the environmental perception ability of the system and reduces the deployment complexity without adding additional dedicated sensor hardware by multiplexing the acoustic function of the distributed execution and perception module; each distributed module is both a cooling actuator and an acoustic detector; this dual-mode design, combined with a wireless sensing sub-network for environmental data collection, enables the system to obtain rich physical environment state information at a lower cost, while significantly simplifying the wiring construction on site, improving the integration and maintainability of the system, and achieving significant economic and social benefits. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 is a schematic diagram of the overall architecture of the system of the present application; Figure 2 is a system overall control flowchart of the present application. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0021] Embodiment: Please refer to the accompanying Figure 1 The present application provides a refrigeration and cold storage Internet of Things intelligent temperature control and energy consumption management system, which comprises a distributed execution and perception module, an environmental data acquisition module, a central processing and decision module, and an energy and communication support module.

[0022] The distributed execution and sensing modules are deployed in an array form in the controlled temperature interior space. The basic constituent unit of the distributed execution and sensing module has dual functions: one is to perform refrigeration tasks based on the thermoacoustic effect, directly pumping heat to the adjacent space; the other is to detect the acoustic response of the interior space under the control of the control instruction, with the acoustic driver inside as the acoustic excitation source and the built-in acoustic pressure sensor as the acoustic signal receiving point.

[0023] The environmental data acquisition module is composed of a plurality of micro-environmental sensors deployed in a spatial gridding manner. These sensors are responsible for collecting and periodically reporting the temperature, humidity and other environmental parameters at their locations, forming a high-density, real-time updated environmental parameter data set.

[0024] The central processing and decision-making module is the computing and control core of the system. It receives environmental parameter data from the environmental data acquisition module, and its own state data and real-time acoustic response data from the distributed execution and sensing module. The module is responsible for running data processing, state evaluation and optimization decision-making algorithms, and finally generating control instructions for each distributed execution and sensing module.

[0025] The energy and communication support module provides the operating power required by all electronic components in the system, and builds a stable data communication network covering all modules, ensuring that control instructions can be reliably issued and data can be uploaded.

[0026] The distributed execution and sensing module is the basic physical unit for realizing refrigeration and environmental acoustic sensing. The module as a whole can be constructed as a flat block or tile structure, facilitating array integration installation on the inner wall, top or shelf of the freezer or refrigerator.

[0027] The distributed execution and sensing module includes a shell with a predetermined geometric shape, which can be made of metal materials with good thermal conductivity or high-strength engineering plastics, for packaging internal components and providing structural support. Inside the shell are integrated acoustic drivers, resonant tubes, thermodynamic cores, i.e. plate stacks, cold-end heat exchangers, hot-end heat exchangers, acoustic pressure sensors, and microcontrollers.

[0028] The acoustic driver is used to convert the input electrical energy into high-intensity reciprocating mechanical vibration, thereby driving the gas medium in the resonant tube to generate sound waves. The acoustic driver can be a moving coil or moving iron linear motor, or a piezoelectric ceramic stack actuator. The selection and structural design of the driver can be selected by those skilled in the art according to the required acoustic power, frequency and efficiency, and the specific implementation belongs to the known technology in the art, which will not be described here.

[0029] The resonant tube is a closed or semi-closed cavity with a specific length and shape. One end of the resonant tube is coupled with the acoustic wave driver, used to contain the working gas and establish a stable standing wave field in it. The material of the resonant tube needs to have good airtightness and acoustic performance.

[0030] The plate stack is arranged at a specific position in the internal acoustic field of the resonant tube. It is composed of a large number of parallel thin plates or ceramic materials with a microporous structure, and extremely narrow fluid channels are formed between the plates. The working gas reciprocally oscillates and exchanges heat in these channels under the action of acoustic waves, thereby realizing a thermodynamic cycle.

[0031] The cold-end heat exchanger and the hot-end heat exchanger are arranged at both ends of the plate stack, respectively. The surface of the cold-end heat exchanger is directly or indirectly exposed to the space inside the library that needs to be refrigerated. Its structure can be designed as a finned or micro-channel type to increase the heat exchange area with the air in the library for absorbing heat from the library. The hot-end heat exchanger is used to dissipate the heat absorbed from the cold-end and pumped by the thermoacoustic effect, as well as the waste heat in the acoustic power conversion process, to the environment outside the library or a specially designed cooling circuit.

[0032] The acoustic pressure sensor is used to detect the acoustic pressure changes inside the resonant tube or the space inside the library. The sensor can be a micro-electro-mechanical system microphone or a piezoelectric acoustic sensor, which is installed on the tube wall of the resonant tube at the position of the pressure wave crest to obtain the highest detection sensitivity. The acoustic signals collected by the sensor are transmitted to the microcontroller for processing.

[0033] The microcontroller is the local control core of the module, usually a single-chip microcomputer or an embedded system processor. It receives instructions from the central processing and decision-making module through the communication interface and controls the operating parameters of the acoustic wave driver according to the instructions. At the same time, it is responsible for reading the data of the acoustic pressure sensor and uploading the data in the detection mode.

[0034] In the embodiment of the present application, the above-mentioned integrated manner of the distributed execution and perception module in the cold storage can be realized by the following steps: a standardized installation interface matching the shell of the distributed execution and perception module is arranged at a predetermined position on the inner wall, ceiling or shelf system of the cold storage. The interface can include a mechanical fixing structure and a line connection port. The body of the distributed execution and perception module is fixed to the installation interface by buckling, bolt fastening, etc., to ensure that its cold-end heat exchanger faces the space inside the library. Through the line connection port on the installation interface, the power input end of the distributed execution and perception module is connected with the direct-current power bus of the energy and communication support module, and its data communication end is connected with the communication network.

[0035] The refrigeration function of the distributed execution and sensing module is realized based on a standing wave thermoacoustic effect, and a core thereof is to convert acoustic work into directional heat pumping. The principle does not involve phase change of a working fluid, and no other macroscopic moving parts are involved in a core thermodynamic process except for an acoustic wave driver.

[0036] A refrigeration working principle based on the thermoacoustic effect can include the following steps: Sound field establishment. A microcontroller controls input electric power and a working frequency of the acoustic wave driver according to a received refrigeration power instruction. The acoustic wave driver converts electric energy into mechanical vibration energy, and the vibration excites working gas in a resonant tube, and when the driving frequency matches a certain order acoustic resonance frequency of the resonant tube, a high-intensity standing wave sound field is established in the tube. The standing wave sound field is characterized in that a pressure wave crest and a velocity wave crest are different by about one-quarter of a wavelength in space. The plate stack is arranged in a region between the pressure wave crest and the velocity wave crest.

[0037] Thermodynamic cycle of the gas micro-cluster. In the micro-channel of the plate stack, the gas micro-cluster close to the plate wall experiences a periodic thermodynamic process similar to an inverse Brayton cycle under the action of the standing wave sound field. The cycle can be divided into four stages: The gas micro-cluster moves from the velocity wave crest to the pressure wave crest, and experiences a compression process, and its pressure and temperature rise.

[0038] After reaching the maximum compression state, the gas micro-cluster is displaced along the plate wall surface. At this time, because its temperature is higher than the temperature of the adjacent plate wall, heat is transferred from the gas micro-cluster to the plate wall.

[0039] The gas micro-cluster moves from the pressure wave crest to the velocity wave crest, and experiences an expansion process, and its pressure and temperature decrease, and at this time, its temperature is lower than the temperature of the adjacent plate wall.

[0040] After reaching the maximum expansion state, the gas micro-cluster reverses the displacement along the plate wall surface in the initial direction. In this process, because its temperature is lower than the temperature of the plate wall, heat is transferred from the plate wall to the gas micro-cluster.

[0041] Heat pumping. The above four stages constitute a complete acoustic cycle. In each cycle, the gas micro-cluster absorbs heat from one place of the plate wall and releases heat at another place. The coordinated action of a large number of gas micro-clusters macroscopically shows that heat is continuously pumped along the plate wall of the plate stack from one end to the other end. The heat transport driven by acoustic work makes one end of the plate stack lower in temperature and become a cold end, and the other end higher in temperature and become a hot end.

[0042] External heat exchange. The cold end of the plate stack is in close contact with the cold end heat exchanger, the temperature of which is lowered accordingly, and exchanges heat with the air in the library through its outer surface, absorbing the heat in the library, thereby achieving refrigeration of the space in the library. At the same time, the hot end of the plate stack is in close contact with the hot end heat exchanger, which transmits the heat pumped and the waste heat in the conversion process to the hot end heat exchanger, which is then dissipated to the external environment or cooling medium.

[0043] In this refrigeration principle, the acoustic power generated by the acoustic driver for driving the thermodynamic cycle can be represented by the following formula: ; wherein: represents the time-averaged acoustic power; is the acoustic first-order pressure; is the acoustic first-order velocity; is the conjugate complex of ; is the cross-sectional area of the resonant tube; represents taking the real part of a complex number; represents the differential area element on the cross-sectional area ;

[0044] The final refrigeration power and the consumed acoustic power have a conversion efficiency relationship, which is affected by various factors such as thermodynamic parameters, fluid properties, and plate stack structure. By controlling the electrical power input to the acoustic driver, the size of the acoustic power can be adjusted, and the refrigeration power can be precisely controlled.

[0045] This dual-mode working principle realizes time-sharing multiplexing and reconfiguration of the functions of the acoustic driver and the acoustic pressure sensor through its internal microcontroller, so that a single physical unit can simultaneously undertake the tasks of refrigeration execution and environmental sensing without the need for additional hardware.

[0046] The specific implementation of this dual-mode working principle can include the following steps: Receiving mode switching instructions. The microcontroller of the distributed execution and sensing module receives instructions from the central processing and decision module through the communication interface. The instructions contain mode selection information, i.e., specifying the module to enter the refrigeration mode or the acoustic detection mode. By default or without special instructions, the module is usually in the refrigeration mode.

[0047] The refrigeration mode is executed. When the module is in refrigeration mode, the microcontroller controls the acoustic wave driver to operate at high power, at a specific acoustic resonance frequency that matches the resonant tube, thus establishing a stable standing wave acoustic field, realizing the aforementioned refrigeration based on thermoacoustic effect. In this mode, the acoustic pressure sensor can be used to monitor the acoustic pressure amplitude in the resonant tube as a feedback signal for local closed-loop control, or be in a dormant state.

[0048] The acoustic probing mode is executed. When the received instruction is to enter the acoustic probing mode, the microcontroller reconfigures the functions of the module. First, the power driving the acoustic wave driver is reduced to a lower level, which is insufficient to produce significant thermodynamic effects, but sufficient to produce an acoustic wave in the space of the library that can be clearly probed. Second, the driving signal waveform of the acoustic wave driver is changed. The driving signal for probing is no longer a single frequency sinusoidal wave for refrigeration, but a specific waveform for probing the acoustic characteristics of the space, a linear frequency modulation signal with a wide frequency band: ; wherein: represents the instantaneous probe signal; is the amplitude of the probe signal; is the starting frequency; is the rate of change of frequency over time, i.e. the frequency modulation rate; is the duration of the probe signal.

[0049] Acoustic response is collected. At the same time as the acoustic wave driver emits the probe signal , the microcontroller of the same module or other modules within a specified range activates its acoustic pressure sensor and collects and digitizes the received acoustic signal at a preset sampling frequency. The collected signal is the acoustic response formed after the probe signal interacts with the entire space of the library, which can be modeled as: ; wherein: is the room impulse response representing the physical environment of the space; represents the convolution operation; represents environmental and circuit noise.

[0050] Data upload and mode recovery. The microcontroller uploads the collected digitized acoustic response data to the central processing and decision module via the communication network for subsequent analysis. After the completion of the acoustic probing task, the microcontroller can restore the distributed execution and perception module to the refrigeration mode or keep it in standby state according to the instruction.

[0051] Through the above-mentioned dual-mode working principle, the application multiplexes the core driving component of the refrigeration system as an excitation source of the acoustic sensing system, multiplexes the internal monitoring sensor as an acoustic receiver, and realizes high integration of hardware and effective expansion of functions.

[0052] To ensure that the distributed execution and perception module can accurately and quickly respond to the control instructions issued by the central processing and decision module, the internal microcontroller performs local and high-speed closed-loop control. The goal of this local closed-loop control is to make the actual refrigeration power it generates stably track the target refrigeration power instruction issued by the central module, so as to convert macroscopic and slow-period global optimization scheduling into microscopic and real-time accurate physical execution.

[0053] The implementation of the local closed-loop control can include the following steps: Receiving instructions and setting target values. The microcontroller receives the optimal operating power instruction from the central processing and decision module through the communication interface. The instruction can be parsed as a specific target refrigeration power . The microcontroller converts this target refrigeration power into a physical quantity corresponding thereto that can be directly measured by the internal sensor as a control set point . The physical quantity can be specifically the sound pressure amplitude of the standing wave acoustic field in the resonant tube , or the surface target temperature of the cold end heat exchanger .

[0054] Obtaining feedback. The microcontroller periodically and at a high frequency reads the measurement value of the internal sensor to obtain the real-time feedback . If the control set point is the sound pressure amplitude , the microcontroller reads the measurement value of the sound pressure sensor, and obtains the actual sound pressure amplitude as the feedback after signal processing.

[0055] Calculating the control error and executing the control algorithm. The microcontroller compares the control set point with the real-time feedback , and calculates the control error .

[0056] ; Subsequently, the microcontroller executes the preset control algorithm in a fixed and high-frequency control period, and calculates the control output quantity of the sound wave driver according to the control error . The control algorithm can be a proportional-integral-derivative controller. In the discrete time domain, the output of the PID control algorithm can be represented as: ; wherein: is the discrete time step, is the control output at the th time step, is the control error at the th time step, is the sampling period of the controller, are the proportional, integral and derivative gain coefficients of the PID controller, respectively. For the parameter setting of the PID controller, one skilled in the art can use the Ziegler-Nichols method or other modern control theory methods.

[0057] The microcontroller will act the calculated control output on the power regulation circuit of the acoustic driver. The control output can be used to adjust the duty cycle of a pulse width modulation signal, which in turn controls a power amplifier that supplies power to the acoustic driver, so as to accurately adjust the actual electrical power input to the acoustic driver.

[0058] Through the rapid circulation of the above steps, the local closed-loop control system can dynamically adjust the input power of the acoustic driver, compensate for performance fluctuations caused by factors such as changes in working gas temperature, external environmental disturbances, etc., and ensure that the actual physical state of the module is stably maintained at the target set point, thereby ensuring that the refrigeration power output by the module accurately meets the scheduling instructions of the central processing and decision-making module.

[0059] An environmental data acquisition module, which is physically manifested as a distributed network composed of multiple independent, low-power sensor nodes. The network aims to obtain the basic environmental parameters required for reconstructing the three-dimensional temperature field and conducting fine management by deploying a large range and high density within the controlled space.

[0060] Each of the sensor nodes includes one or more sensing units, a micro-processing unit, a wireless communication unit, and an independent power supply unit.

[0061] The sensing unit is used to directly measure the environmental physical quantity at the location of the node. The sensing unit can include a temperature sensing element for measuring temperature and a humidity sensing element for measuring relative humidity. The temperature sensing element can be a negative temperature coefficient thermistor, a platinum resistance, or an integrated digital temperature sensor chip. The humidity sensing element can be a capacitive or resistive humidity sensor.

[0062] The micro-processing unit, usually a low-power microcontroller, is responsible for waking up the system at preset time intervals, controlling the sensor unit to collect data, and reading the analog or digital signals output by the sensor unit. After reading, the micro-processing unit can perform preliminary processing on the raw data, analog-to-digital conversion on the analog signals, digital filtering on the data to reduce noise, and encapsulate the processed temperature and humidity data into data packets conforming to the predetermined communication protocol.

[0063] The wireless communication unit is responsible for sending the data packets processed by the micro-processing unit to the central processing and decision-making module in a wireless manner. The wireless communication unit can be a wireless transceiver chip supporting a low-power wide-area network protocol or a wireless transceiver chip supporting a near-field communication protocol. The choice of wireless communication protocol can be made by the skilled person in the art based on the specific size and structure of the cold storage, as well as the required communication distance and power consumption.

[0064] The independent power supply unit is used to provide power for all electronic components on the sensor node. The power supply unit is a long-life battery, such as a disposable lithium sulfonyl chloride battery or a rechargeable lithium-ion battery, ensuring that the sensor node can operate autonomously for a long time without external wiring. The firmware design of the micro-processing unit adopts a low-power strategy, allowing itself and the peripheral circuit to enter a deep sleep mode during the interval between two collection intervals, thereby maximizing the battery life.

[0065] To adapt to the low-temperature and high-humidity environment of the freezer and refrigerator, the above-mentioned sensor unit, micro-processing unit, wireless communication unit, and power supply unit are integrated on a printed circuit board and encapsulated in a sealed housing with a predetermined protection level to prevent water vapor from entering and causing damage to electronic components.

[0066] To ensure that the data collected by the environmental data collection module accurately reflects the temperature distribution state of the three-dimensional space inside the library, the embodiment of the present application adopts a structured sensor node deployment strategy and specifies a corresponding data transmission mechanism.

[0067] The deployment strategy of the sensor node can be implemented by the following steps: Spatial gridding. Before deployment, the internal three-dimensional size of the cold storage and the required temperature field monitoring accuracy are used to logically divide it into a three-dimensional virtual grid. The node or center point of each grid is the predetermined sensor deployment location. The density of the grid, i.e., the distance between adjacent deployment locations, can be adjusted according to application requirements. For areas requiring high-precision temperature control, a denser grid deployment can be used.

[0068] Node physical deployment. According to the results of the spatial gridding described above, the sensor nodes are physically installed at the corresponding predetermined locations. The deployment locations are not limited to the inner walls and ceiling of the warehouse. The sensor nodes are also directly fixed on the storage facilities in the warehouse, the columns or beams of the shelves. For scenarios that need to monitor the temperature of specific goods, the sensor nodes can also be placed on the pallets or stored with the goods. In addition, additional sensor nodes should be added for key monitoring at some thermodynamic key locations, near the door, the outlet of the fan, and the cold end of each distributed execution and perception module.

[0069] After the sensor nodes are deployed, the process of transmitting their data to the central processing and decision module is as follows: Data packet formatting. After collecting the temperature and humidity data, the microprocessor unit of each sensor node encapsulates the data into a standard data packet along with its unique identifier, the collection timestamp, and the current state of the power supply unit, etc. The data packet can further contain a check code to ensure the integrity of the data during transmission.

[0070] Data wireless transmission and aggregation. The sensor nodes broadcast or point-to-point send the encapsulated data packets through their wireless communication units. One or more wireless gateways deployed in or near the cold storage are responsible for receiving wireless signals from various sensor nodes. The function of the wireless gateway is to convert the low-power, short-range wireless communication protocol into a high-bandwidth, long-range network communication protocol.

[0071] Data routing to the central module. After receiving and verifying the data packets, the wireless gateway routes and transmits the aggregated sensor data to the database or real-time data stream interface of the central processing and decision module through wired or wireless means. In this way, the central processing and decision module can continuously obtain a complete set of structured environmental data with spatial location and timestamp information, providing a data basis for subsequent three-dimensional temperature field reconstruction and system decision-making.

[0072] The data fusion and three-dimensional temperature field reconstruction unit inside the central processing and decision module is a key link connecting the bottom layer of data collection and the upper layer of optimization decision-making. The function of this unit is to receive heterogeneous data from various modules and generate a continuous three-dimensional temperature field model that can represent the thermodynamic state of the entire controlled space based on the environmental parameter data.

[0073] The working process of this unit can include the following steps: Multimodal data receiving and preprocessing. This unit receives the temperature and humidity data stream from the environmental data acquisition module, and the device status data stream from the distributed execution and perception module. Firstly, this unit time-aligns the data from different sources according to the timestamp information in each data packet, forming a synchronized data set. Secondly, the received data is subjected to validity test, and obvious outliers caused by sensor failure or communication error are eliminated by setting reasonable physical quantity threshold, and the missing data points in a short time are interpolated and filled with the data of adjacent time.

[0074] Three-dimensional temperature field reconstruction. The core task of this unit is to calculate the estimated temperature at any target point by spatial interpolation algorithm, using the preprocessed, spatially discrete distributed temperature measurement values . Inverse distance weighted interpolation method is used for reconstruction. The calculation formula of this method is: ; wherein the weight is calculated as: ; Here: is the estimated temperature at time , spatial coordinate ; is the temperature value measured by the th sensor at time , whose spatial coordinate is ; is the Euclidean distance from the target point to be interpolated to the position of the th sensor; is a positive real power index, used to adjust the decay rate of the weight with distance, usually taking the value of 2.

[0075] In order to obtain higher interpolation accuracy, Kriging interpolation method can also be used. This method not only considers the influence of distance, but also analyzes the spatial autocorrelation between the measurement values of each sensor by using the variogram function, which can provide optimal linear unbiased estimation.

[0076] Model output. This unit logically divides the entire controlled space into a fine three-dimensional voxel grid, and calculates the temperature value of each voxel center using the above interpolation algorithm. Finally, this unit outputs the discretized three-dimensional temperature field model in the form of a three-dimensional array at the current time . This model is passed to the subsequent functional units for calculation and visualization of local cooling demand.​​

[0077] The acoustic state inversion unit in the central processing and decision-making module is responsible for in-depth analysis of acoustic response data uploaded from the distributed execution and sensing module. Its core function is to reverse-analyze changes observed at the acoustic level into changes in the state of the physical environment within the controlled space. This unit is the technical foundation for realizing the forward-looking control and abnormal state monitoring of this invention.

[0078] The operation of this acoustic state inversion unit may include the following steps: The unit calculates the chamber impulse response and constructs the acoustic deviation vector. It receives real-time acoustic response signals acquired in acoustic detection mode. It is known that this signal is a probe signal. With chamber impulse response The convolution result. This unit first performs a deconvolution operation from... and known Solve for the current chamber impulse response. This deconvolution operation can be implemented using signal processing methods such as Wiener filtering. Subsequently, the calculated current chamber impulse response will be... Compared with the acoustic environment reference model established in step S1 and stored in the system, i.e., the reference chamber impulse response A comparative analysis was conducted, and the differences were quantified into a multi-dimensional acoustic deviation vector. This vector may include the following components: Resonant frequency offset component: through the and Perform Fourier transforms on each frequency spectrum to obtain its corresponding spectrum. Determine the current principal resonant frequency by searching for peak points in the spectrum. and reference resonant frequency And calculate its offset. .

[0079] The decay time variation component: Based on methods such as Schroeder's integral method, respectively from... and Calculate the reverberation time of the space and And calculate its change. .

[0080] Energy attenuation spectrum variation component: Divide the spectrum of the two impulse responses into multiple frequency bands and calculate the energy difference within each frequency band.

[0081] The final acoustic deviation vector is .

[0082] Inverse analysis of cargo loading status based on acoustic deviation vector. This step involves constructing the acoustic deviation vector... to infer the loading state of the goods in the warehouse. This inverse analysis function is realized by a pre-trained machine learning model. The construction and use of this model is as follows: in the early stage of system deployment or maintenance, a variety of known and representative loading states of goods are artificially set in the cold storage, and for each state, the acoustic probe is run and the corresponding acoustic deviation vector is calculated . These loading states are used as labels, and the corresponding acoustic deviation vectors are used as features to form a training data set.

[0083] Using this data set, a classification or regression model is trained, such as a support vector machine, a decision tree, or a small neural network. During normal operation of the system, the unit will calculate the acoustic deviation vector in real time as input into the trained model, and the output of the model is the estimated result of the current loading state of the goods in the warehouse. When the resonant frequency component in the acoustic deviation vector decreases systematically, the model can infer that the effective volume in the warehouse has decreased, i.e. the loading rate has increased.

[0084] Monitoring of gas composition anomalies based on acoustic deviation vectors. This step uses the acoustic deviation vector to monitor changes in the average molar mass of the gas in the warehouse due to spoilage of goods, etc. The physical basis of this monitoring is the relationship between sound velocity and the properties of the gas medium. In the ideal gas model, the sound velocity can be given by the following formula: ; where: is the adiabatic index of the gas; is the universal gas constant; is the absolute temperature of the gas; is the average molar mass of the gas.

[0085] The specific monitoring process of this unit is as follows: first, the main resonant frequency shift is extracted from the acoustic deviation vector . Since the resonant frequency of the space is directly proportional to the sound velocity , this frequency shift directly reflects the change in sound velocity. Second, the average temperature in the warehouse is obtained from the data fusion and three-dimensional temperature field reconstruction unit. Based on the sound velocity and temperature of the reference state, the sound velocity that should exist at the current temperature if the gas composition has not changed is calculated. Then, based on the current actual measured resonant frequency , the actual sound velocity is calculated. If and a deviation between the average molar mass of the gas and a preset threshold value is more than a preset threshold value, indicating that the average molar mass of the gas is changed. The unit judges that the gas composition in the warehouse is abnormal and can generate an alarm signal.

[0086] The feedforward optimization decision unit inside the central processing and decision module is the calculation core of the energy efficiency optimization of the present application. The responsibility of the unit is to fuse the current state information of the system and the future heat load prediction based on acoustic perception, establish and solve a mathematical optimization problem with the minimum global energy consumption as the target, and finally determine the optimal control strategy for each distributed execution and perception module.

[0087] The working process of the feedforward optimization decision unit can include the following steps: Fuse the total refrigeration demand. The unit receives the real-time three-dimensional temperature field model from the data fusion and three-dimensional temperature field reconstruction unit, and calculates the reaction refrigeration demand based on this. The unit divides the space in the warehouse into control areas. For the first control area, the reaction refrigeration demand is calculated by comparing the average temperature of the area with the preset target temperature . At the same time, the unit receives the physical environment state evaluation results from the acoustic state inversion unit. If the evaluation results show that new, uncooled goods enter, the unit calculates the prospective heat load prediction value, i.e. the forward-looking refrigeration demand , according to the estimated mass, specific heat capacity and initial temperature of the goods. Then, the unit fuses the two demands to obtain the total refrigeration demand of the first control area: ; wherein and are preset weight coefficients for adjusting the proportion of the two demands in decision-making.

[0088] Build a global energy consumption optimization model. The unit minimizes the sum of the total input electric power of all distributed execution and perception modules in the system as the optimization target, and establishes a multivariate constraint optimization model. The model can be expressed as: Objective function: ; Constraint condition: Refrigeration demand constraint: for each control area , the total refrigeration amount provided by all modules that have a refrigeration effect on it must be greater than or equal to the total refrigeration demand of the area.

[0089] ; Module capability constraint: each module must be within its minimum cooling power and maximum cooling power .

[0090] ; where: is the cooling power to be solved for the th module; is the electrical power consumed by the th module to produce cooling power ; is the electrical to cooling power conversion function of the th module, which is measured in the system initialization calibration stage and can be a polynomial function or a look-up table; is the influence coefficient, which represents the contribution of the cooling power produced by the th module to the temperature of the th control zone. This coefficient is related to the relative spatial position of the module and the zone.

[0091] Solve the optimization problem. This unit calls a numerical optimization algorithm to solve the optimization model constructed above. Since the function is usually nonlinear, this optimization problem is a nonlinear programming problem. Sequential quadratic programming, interior point method, etc. can be used to solve it. The result of the solution, i.e. a set of optimal cooling power allocation schemes that can minimize the total energy consumption of the system while meeting all the cooling demands of the zones , is passed to the instruction generation and scheduling unit.

[0092] The instruction generation and scheduling unit inside the central processing and decision-making module is the final output link connecting the decision-making layer and the execution layer in the entire control loop. This unit receives the calculation results from the feedforward optimization decision-making unit and converts them into physically executable and specific device control instructions, which are then issued to the corresponding distributed execution and perception modules through the communication network.

[0093] The working process of this instruction generation and scheduling unit can include the following steps: Analyze the optimization results and generate cooling instructions. This unit receives the optimal cooling power allocation scheme from the feedforward optimization decision-making unit. For each module, this unit converts the abstract cooling power target value Translates to physical quantity set points that can be directly used by the internal local partial closed-loop controllers. The unit looks up in a stored look-up table or polynomial function the corresponding target input electrical power or target sound pressure amplitude . .

[0094] Probe instruction generation and scheduling. In addition to cooling instructions, this unit is also responsible for generating and scheduling acoustic probe tasks. According to pre-set time strategies or event-triggered strategies, this unit initiates an acoustic probe proactively. At this time, the unit generates a set of probe instructions, specifying one or more distributed execution and perception modules as sound sources, and other modules as acoustic receivers. The probe instructions contain mode switching information, probe signal parameters, etc.

[0095] Instruction data packet formatting. This unit encapsulates the above-mentioned generated cooling instructions or probe instructions, together with the unique identifier of the target module, into a structured instruction data packet. The data packet follows the internal communication protocol defined by the system, and its format can include the following fields: target address field, source address field, instruction type field, instruction data payload field, and cyclic redundancy check code field for data verification.

[0096] Instruction delivery. This unit sends the formatted instruction data packet to the energy and communication support module through the internal bus. The energy and communication support module then routes the instruction data packet to the specified distributed execution and perception module through wired or wireless networks according to the target address field in the data packet. After receiving and verifying the data packet, the microcontroller of the target module parses the instructions and performs corresponding local closed-loop control or mode switching operations.

[0097] The energy and communication support module provides basic energy supply and management for the entire system. The system adopts a centralized management direct current power bus architecture to provide stable and efficient electrical energy for all distributed distributed execution and perception modules, central processing and decision-making modules, and wireless gateways, etc. active devices.

[0098] The specific implementation of the direct current power bus and power supply management can include the following steps: Centralized power conversion. The system sets up one or more centralized power conversion units. The unit accesses the external AC power grid and converts it into stable, relatively safe DC voltage for personnel through a high-power switching power supply. The DC voltage is fed to the main road of the DC power bus. Compared with configuring a converter for each remote device, this centralized conversion and DC power supply method can reduce the overall AC-DC conversion loss of the system and improve energy utilization efficiency.

[0099] DC bus layout. The DC power bus is composed of wires with sufficient current-carrying capacity and low resistance. The main trunk and branch lines are laid along the walls, roof or dedicated wire slots of the cold storage. The wiring path is planned to ensure that power access points can be conveniently provided for each installation location of the distributed execution and perception modules.

[0100] Power supply management and protection. A power management unit is provided in the system to monitor and protect the state of the DC power bus in real time. Specifically, the power management unit integrates a Hall effect current sensor and a voltage sensor on the main trunk of the DC bus to continuously measure the total output current and bus voltage of the bus. The power management unit also has overcurrent protection, short circuit protection and overvoltage / undervoltage protection functions. When an abnormal increase in total current or a bus voltage beyond the preset safety range is detected, the unit can control the circuit breaker or solid-state relay on the main circuit to quickly cut off the power output to protect the safety of all devices in the system.

[0101] Electricity data reporting. The power management unit processes the real-time monitored total current and bus voltage data to calculate the total power consumption of the system at the current time. This total power consumption data is periodically reported to the central processing and decision module through the communication network. The central processing and decision module can compare the actual measured total power with the theoretical total power calculated by the feedforward optimization decision unit in the model to evaluate the accuracy of the optimization model or diagnose whether there is abnormal energy consumption in the system.

[0102] The energy and communication support module provides a hierarchical and efficient data exchange channel for different types of devices in the system. To balance the different needs of different devices in terms of communication bandwidth, power consumption and deployment flexibility, the system adopts a hybrid communication network architecture. This architecture integrates a high-speed wired backbone network and a low-power wireless sensor subnetwork.

[0103] The specific implementation of the hybrid communication network architecture can include the following steps: Build a wired backbone network. This network is responsible for connecting the core devices in the system that have large data throughput and require high real-time and reliability of communication. The wired backbone network uses industrial Ethernet technology. The central processing and decision module, the wireless gateway that provides access to the wireless sensor subnetwork, and all distributed execution and perception modules are connected to the industrial-level network switch deployed inside the energy and communication support module through Ethernet cables. This connection ensures low-latency delivery of control instructions and fast uploading of large data generated by tasks such as acoustic detection.

[0104] A wireless sensing sub-network is constructed. The network is specially used to connect a large number of low-power environmental data collection modules which are deployed in space and powered by batteries. The wireless sensing sub-network adopts Zigbee protocol to construct a self-organizing mesh network. Each environmental data collection module acts as a terminal node in the network, and sends the temperature and humidity data collected by it to the nearest parent node or routing node. The data is finally gathered to the wireless gateway acting as the network coordinator through multi-hop routing. The use of such a wireless network greatly simplifies the deployment of a large number of sensors and avoids complex wiring engineering. For the selection of wireless communication protocol and network implementation, those skilled in the art can configure according to the communication distance, node density and cost.

[0105] Network protocol conversion and data aggregation are implemented. The wireless gateway is a bridge connecting the above-mentioned wired backbone network and the wireless sensing sub-network. Its core function is protocol conversion. The wireless gateway receives and analyzes the data packets from the wireless sensing sub-network which comply with a specific low-power wireless protocol through its wireless radio unit, then re-encapsulates the payload in the data packets into Ethernet data frames complying with TCP / IP protocol, and sends them to the wired backbone network through its Ethernet interface. Conversely, configuration instructions for wireless nodes from the central processing and decision module can also be received and converted into wireless signals for distribution.

[0106] Through this hybrid architecture, the system provides stable wired connection for the execution module which needs high bandwidth and high reliability, and convenient wireless connection for the sensing node which needs low power consumption and high flexibility, achieving a balance in performance, cost and scalability of the entire communication system.

[0107] Please refer to the attached Figure 2 A distributed intelligent temperature control and energy consumption management method based on thermoacoustic effect and sound field perception includes the following steps: S1, acoustic response data and temperature and humidity data in the space are respectively acquired by the distributed execution and perception module and the environmental data collection module; S2, the physical environment state in the space is inverted based on the acoustic response data; S3, the optimization model with the minimum total energy consumption of the system as the target is established and solved by combining the physical environment state and the temperature and humidity data, and control instructions for each distributed execution and perception module are generated; S4, the control instructions are issued to drive the distributed execution and perception module to perform refrigeration operation.

[0108] The method of the embodiment can be used to implement the above-mentioned system embodiment, and the principles and technical effects are similar, which will not be repeated here.

Claims

1. A smart temperature control and energy consumption management system for refrigeration and freezing IoT, characterized in that, include: Multiple distributed execution and sensing modules, each of which is configured to provide cooling capacity in cooling mode and to transmit and receive sound waves to acquire acoustic response data in acoustic detection mode; Multiple environmental data acquisition modules are used to collect temperature and humidity data of the space where the system is located; The central processing and decision-making module is communicatively connected to the distributed execution and sensing module and the environmental data acquisition module, respectively. The central processing and decision-making module is used to receive the temperature and humidity data and the acoustic response data, invert the physical environment state in the space based on the acoustic response data, and combine the temperature and humidity data to establish an optimization model with the goal of minimizing the total energy consumption of the system, and determine the control instructions for each of the distributed execution and sensing modules. An energy and communication support module is used to provide energy supply and communication support to the system.

2. The intelligent temperature control and energy consumption management system for refrigeration and freezing IoT as described in claim 1, characterized in that, The distributed execution and sensing module is a thermoacoustic cooling module, specifically including: an acoustic wave driver for generating sound waves, a resonant tube for acoustic wave resonance, a stack for generating a temperature difference in the sound field, a cold end heat exchanger located on the low-temperature side of the stack, and a sound pressure sensor for measuring the acoustic response in the acoustic detection mode.

3. The intelligent temperature control and energy consumption management system for refrigeration and freezing IoT as described in claim 1, characterized in that, The central processing and decision-making module specifically includes: A data fusion and three-dimensional temperature field reconstruction unit is used to generate a three-dimensional temperature field model based on the temperature and humidity data. An acoustic state inversion unit is used to invert the physical environment state based on the acoustic response data; A feedforward optimization decision unit is used to establish and solve the optimization model based on the evaluation results of the three-dimensional temperature field model and the physical environment state. The instruction generation and scheduling unit is used to convert the solution results of the optimization model into the control instructions and issue them.

4. The intelligent temperature control and energy consumption management system for refrigeration and freezing IoT as described in claim 3, characterized in that, The acoustic state inversion unit specifically includes: obtaining an acoustic deviation vector by comparing the real-time chamber impulse response with a preset reference chamber impulse response, and inversely analyzing the cargo loading status or gas composition anomaly in the space based on the acoustic deviation vector, as the evaluation result of the physical environment state.

5. The intelligent temperature control and energy consumption management system for refrigeration and freezing IoT as described in claim 3, characterized in that, The feedforward optimization decision unit specifically includes: integrating the reactive cooling demand calculated based on the three-dimensional temperature field model and the forward-looking cooling demand calculated based on the physical environment state assessment results output by the acoustic state inversion unit to generate the overall cooling demand for the optimization model.

6. The intelligent temperature control and energy consumption management system for refrigeration and freezing IoT as described in claim 5, characterized in that, The optimization model established by the feedforward optimization decision unit includes the following constraints: cooling demand constraints that meet the overall cooling demand, and module capability constraints for each of the distributed execution and sensing modules.

7. The intelligent temperature control and energy consumption management system for refrigeration and freezing IoT as described in claim 1, characterized in that, The energy and communication support module provides a hybrid communication network architecture, which includes a wired backbone network for connecting the central processing and decision-making module and the multiple distributed execution and sensing modules, and a wireless sensing subnetwork for connecting the one or more environmental data acquisition modules.

8. The intelligent temperature control and energy consumption management system for refrigeration and freezing IoT as described in claim 1, characterized in that, The energy and communication support module also includes a DC power bus, which is used to centrally convert external AC power into DC power and to power the multiple distributed execution and sensing modules and the central processing and decision-making module.

9. A smart temperature control and energy consumption management system for refrigeration and freezing IoT as described in claim 3, characterized in that, The data fusion and three-dimensional temperature field reconstruction unit specifically includes: using the spatially discrete temperature and humidity data, and through a spatial interpolation algorithm, generating a three-dimensional temperature field model of the space.

10. A method for intelligent temperature control and energy management of frozen and refrigerated storage using the Internet of Things (IoT), as described in any one of claims 1-9, characterized in that, Includes the following steps: S1. Acquire acoustic response data and temperature and humidity data within the space through the distributed execution and sensing module and the environmental data acquisition module, respectively; S2. Based on the acoustic response data, invert the physical environment state within the space; S3. Combining the physical environment status and the temperature and humidity data, establish and solve an optimization model with the goal of minimizing the total energy consumption of the system, and generate control instructions for each of the distributed execution and sensing modules; S4. Issue the control command to drive the distributed execution and sensing module to perform the cooling operation.

Images