Temperature and humidity control system and method for clinical medicine refrigerated cabinet and refrigerated cabinet

Abstract

The invention belongs to the technical field of temperature and humidity regulation and control, and discloses a clinical medicine refrigerated cabinet temperature and humidity control system and method and a refrigerated cabinet, and the method comprises the steps: carrying out the data collection of temperature and humidity related parameters in a target refrigerated cabinet, and obtaining the corresponding temperature and humidity data; data analysis is carried out on the collected temperature and humidity data, whether the corresponding temperature and humidity data are abnormal or not is judged based on a data analysis result, if yes, corresponding regulation and control instructions are generated, and the regulation and control instructions comprise a temperature regulation and control instruction and a humidity regulation and control instruction; obtaining required regulation power based on the obtained regulation instruction, and generating a corresponding regulation signal based on the required regulation power; adjusting the temperature and humidity data of the corresponding refrigerated cabinet based on the obtained regulation and control signal, and feeding back information based on an adjustment result; accurate control over the temperature and humidity in the refrigerated cabinet is achieved, and storage safety of clinical medicine is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of temperature and humidity management, and more specifically, to a temperature and humidity control system, method and refrigerator for a clinical drug refrigerator. Background Art

[0002] In the medical field, proper storage of clinical drugs is crucial to ensure their efficacy and quality. Many drugs have strict requirements on the temperature and humidity conditions of the storage environment. Even a slight deviation may cause the drug to deteriorate and become ineffective, endangering the health and medical safety of patients. Traditional refrigerators have many limitations in temperature and humidity control.

[0003] Compared with the existing technology, the traditional refrigerator temperature and humidity control, on the one hand, has a single and simple data collection method, relying only on a small number of sensors for fixed-point monitoring, which makes it difficult to accurately and comprehensively reflect the actual distribution of temperature and humidity in the cabinet, resulting in a lack of precise basis for control instructions, which can easily lead to local temperature and humidity out of control and greatly increase the risk of drug stability damage; for example, the degradation of active ingredients of certain drugs is accelerated outside a specific temperature range, and the uneven temperature and humidity of traditional refrigerators can easily lead to this situation frequently.

[0004] On the other hand, the control mechanism is rigid and inefficient, and the power adjustment is mostly fixed. The power is not accurately adapted according to the real-time temperature and humidity dynamics, and is often over- or under-regulated, wasting energy and aggravating equipment loss, extending the response time, affecting the service life and reliability of the refrigerator, and unable to meet the long-term stable storage needs of drugs. Furthermore, the equipment adjustment process lacks collaborative optimization and intelligent feedback mechanisms, the main and auxiliary compressors have poor coordination, humidity adjustment is delayed or unbalanced, and temperature and humidity fluctuate frequently, seriously threatening the stability of drug quality. There is no effective feedback mechanism to continuously optimize the system, and it is unable to adapt to changes in drug types, storage volume and environment. Frequent manual intervention is costly, inefficient and error-prone.

[0005] In view of this, the present invention proposes a clinical medicine refrigerator temperature and humidity control system, method and refrigerator to solve the above problems. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solutions: A temperature and humidity control system for a clinical drug refrigerator, comprising: A data acquisition module is used to collect data on temperature and humidity related parameters in the target refrigerator to obtain corresponding temperature and humidity data; A data analysis module is used to analyze the collected temperature and humidity data, and determine whether the corresponding temperature and humidity data are abnormal based on the data analysis results. If abnormal, a corresponding control instruction is generated, and the control instruction includes a temperature control instruction and a humidity control instruction; A power acquisition module, which acquires the required control power based on the obtained control instruction and generates a corresponding control signal based thereon; The equipment control module adjusts the temperature and humidity data of the corresponding refrigerator based on the obtained control signal, and provides information feedback based on the adjustment result.

[0007] Furthermore, the process of collecting data on temperature and humidity related parameters in the target refrigerator and obtaining corresponding temperature and humidity data includes: The data collection node is provided with a plurality of collection nodes and identification terminals; based on the collection nodes, data is collected on the temperature, humidity and pressure parameters at different points in the target refrigerator, and the data are transmitted to the corresponding identification terminal, and the identification terminal performs parameter identification and analysis on the received temperature and humidity parameters to obtain corresponding temperature and humidity data and water vapor pressure data.

[0008] Furthermore, the process of performing data analysis on the collected temperature and humidity data and determining whether the corresponding temperature and humidity data is abnormal based on the data analysis result includes: Construct a temperature and humidity change curve; and set a monitoring period, obtain the temperature and humidity data in the current monitoring period based on the temperature and humidity change curve, and obtain the temperature and humidity enthalpy value corresponding to the target refrigerator in the current monitoring period based on the temperature and humidity change curve; Based on the obtained temperature and humidity enthalpy values, an enthalpy value change curve corresponding to the corresponding monitoring period is obtained; Based on the enthalpy change curve, the enthalpy fluctuation coefficient corresponding to the temperature and humidity enthalpy value in the current monitoring period is obtained. ; Obtain the temperature and humidity enthalpy values ​​corresponding to adjacent historical monitoring cycles under normal refrigeration conditions and record them as ; The obtained temperature and humidity enthalpy and Perform a comparison, and determine whether the corresponding temperature and humidity data are abnormal based on the comparison results; like , no other operations are performed; if , the temperature and humidity data in the current monitoring period are marked as abnormal; is the data threshold; Obtain a corresponding temperature data curve based on the temperature and humidity data after abnormal marking; and perform curve analysis on the obtained temperature data curve to obtain a corresponding temperature state and temperature trend, wherein the temperature state includes a high temperature state, a low temperature state and a normal state, and the temperature trend includes a constant temperature trend, a temperature increase trend and a temperature decrease trend; The build interval is A local window is selected, and the current moment is used as the right edge point of the corresponding local window, and the corresponding temperature data curve is intercepted to obtain the corresponding local curve, and the temperature state and temperature trend corresponding to the corresponding local curve are obtained; is a fixed constant; Based on the temperature state and temperature trend corresponding to the local curve, it is judged whether the current temperature needs to be adjusted. If necessary, a corresponding temperature control instruction is generated. If not, a corresponding humidity control instruction is generated.

[0009] Furthermore, the process of acquiring the required control power based on the obtained control instruction and generating a corresponding control signal based on the control power includes: The control power includes temperature control power and humidity control power; the control signal is a temperature control signal or a humidity control signal; Identify the obtained control instruction, and when the temperature control instruction is identified, obtain the corresponding temperature control power based on the temperature data corresponding to the current monitoring period and in combination with the preset adaptive temperature adjustment function, and generate the corresponding temperature control signal based on the temperature data; When the identified control instruction is a humidity control instruction, an adjustment cycle is set, and humidity data within the corresponding adjustment cycle is obtained based on the temperature and humidity data curve, and the optimal humidity parameters corresponding to each moment in the corresponding adjustment cycle are obtained based on the pre-set humidity adaptive function. ; The mean square error values ​​of the corresponding humidity data and the optimal humidity parameters are respectively obtained, and the power compensation coefficient of the corresponding auxiliary compressor is obtained based thereon, the corresponding first output power is adjusted based on the obtained power compensation coefficient, the corresponding humidity control power is obtained, and the corresponding humidity control signal is generated based thereon; wherein the first output power is the real-time output power or rated power of the auxiliary compressor in the refrigerator.

[0010] Furthermore, the formula of the adaptive temperature adjustment function is: P wd = K _p [e(t)+ 1 T_ I ∫ 0 t e( t )d t + T _D de(t) dt ] ; In the formula, , and They respectively represent the best proportional factor, the best integral time factor and the best differential time factor obtained in advance; Indicates the required temperature control power; ; represents the temperature error parameter, and Respectively represent the collected temperature data and the expected temperature data; Represents a time index; The humidity adaptive function is defined as: ; In the formula, is the gas constant for dry air; is the Kelvin temperature corresponding to the current temperature data; is the temperature data at the current moment; Indicates the temperature and humidity enthalpy deviation between the initial moment and the current moment in the adjustment cycle; represents the atmospheric pressure, Indicates the water vapor pressure data in the refrigerator at the current moment; Indicates the volume of the refrigerator.

[0011] Furthermore, the process of obtaining the optimal proportional factor, the optimal integral time factor, and the optimal differential time factor includes: Obtain the temperature data in the current monitoring period and input it into the pre-built parameter control model to obtain the corresponding model output results, which include the required optimal proportional factor, optimal integral time factor and optimal differential time factor; The construction process of the corresponding parameter control model is: The backbone network of the parameter control model is defined as an improved BP neural network, wherein the basic structure of the improved BP neural network is an input layer, a hidden layer and an output layer; The input layer is used to receive input temperature data, obtain corresponding deviation temperature data based on the input temperature data, and construct a corresponding error image based on the deviation temperature data; The hidden layer is used to receive the error image output by the input layer, perform a convolution operation on it, and learn the nonlinear relationship between the input and the output; The formula for the convolution operation is: ; In the formula, For the The hidden layer The output vector of each neuron node; Indicates The output vector of the previous hidden layer; Represents the index of the neuron node in the hidden layer; is the total number of neuron nodes in the corresponding hidden layer, represents the bias term; Indicates The hidden layer The neuron nodes in the previous layer The connection weights between neuron nodes; is a natural number, and For the The index and total number of neuron nodes in the previous hidden layer; represents the activation function; The output layer is used to receive the output vector of the hidden layer and map it into required control parameters, wherein the control parameters include an optimal proportional factor, an optimal integral time factor, and an optimal differential time factor; Acquire temperature data corresponding to a number of historical monitoring cycles and their corresponding temperature control powers; and construct a corresponding training data set based thereon; the training data set is composed of a number of training samples; Initialize the corresponding improved BP neural network, and input the corresponding training samples into the improved BP neural network after the initialization, the improved BP neural network performs forward propagation operation with input layer, hidden layer and output layer, and obtains the corresponding model loss error, compares the obtained model loss error with the preset expected loss error, if the model loss error is less than the expected loss error, then input a new training sample to continue the forward propagation operation, and so on; if the model loss error is not less than the expected loss error, then perform back propagation operation with input layer, hidden layer and output layer as the order, and update the connection weights of the hidden layer and the output layer in the corresponding improved BP neural network based on the preset learning algorithm; The formula for updating weights is: ; In the formula, and They represent the parameters controlling the model in the iterative training process. Second and The adjustment amount corresponding to the secondary connection weight; represents differential operation; and Represent the learning rate and smoothing factor respectively; ) indicates the Deviation temperature data during model iteration training; After the update is completed, repeat the corresponding forward propagation and back propagation operations for iterative training, and record the corresponding model loss error until the obtained model loss error is less than the expected loss error and the model loss error of several consecutive training samples tends to converge, then the model training is completed.

[0012] Furthermore, the process of initializing the corresponding improved BP neural network includes: Obtaining the total number of neuron nodes in the corresponding improved BP neural network, and obtaining the network model parameters required for optimization based on the total number of neuron nodes; and constructing the corresponding fitness function based on the network model parameters; Construct an initial particle swarm based on the network model parameters to be optimized, and obtain the corresponding particle swarm parameters; Obtain the initial velocity and initial position of each particle in the corresponding initial particle group, calculate the fitness of each particle based on the pre-constructed fitness function, and obtain the global optimal solution in the corresponding initial particle group, which is recorded as ; The corresponding initial particle group is randomly divided to obtain several particle groups; the optimal fitness particles in the corresponding particle groups are obtained respectively. and the worst fitness particle ; The worst fitness particle in the corresponding particle group Perform local updates to obtain the corresponding worst fitness particle ; The particle with the worst fitness The particle with the worst fitness Compare, if the particle with the worst fitness Better than the worst fitness particle , then use the particle with the worst fitness Replace the original worst fitness particle ; If the worst fitness particle Not better than the worst fitness particle , then based on the global optimal solution Replace the best fitness particle , and perform local search again to obtain new particles with poor fitness , and compare it with the particle with the worst fitness Perform a second comparison, if it is still not better than the particle with the worst fitness , then a fitness particle is randomly generated to replace the original worst fitness particle And so on, until the number of local updates reaches L times, where L is a fixed constant; After the local update is completed, the updated particle groups are summarized to obtain a new initial particle group, which is randomly grouped again, and the corresponding local update process is repeated to perform mixed iterative training on the corresponding initial particle group until it meets the preset stop condition; Based on the initial particle swarm after hybrid iterative training, the required optimal network model parameters are obtained and applied to the corresponding improved BP neural network.

[0013] Furthermore, the process of adjusting the temperature and humidity data of the corresponding refrigerator based on the obtained control signal and providing information feedback based on the adjustment result includes: Identifying the received control signal, wherein the control signal includes a temperature control signal and a humidity control signal; If the control signal is a temperature control signal, the obtained temperature control power is used as the output power to control the operation of the main compressor in the refrigerator, and the temperature data of the corresponding refrigerator is adjusted. At the same time, based on the acquisition process of the humidity control power, the required real-time humidity control power is obtained, and based on it, the operation of the auxiliary compressor in the refrigerator is controlled, and the humidity data of the corresponding refrigerator is synchronously adjusted until the real-time temperature in the refrigerator is within the expected temperature range and the temperature data is in a constant temperature state; When the temperature data is in a constant temperature state, multiple adjustment cycles are constructed to update the corresponding humidity control power until the corresponding real-time humidity value is within the expected humidity range and is the optimal humidity value corresponding to the current temperature; then the operation of the corresponding auxiliary compressor is stopped; If the control signal is a humidity control signal, the above humidity data adjustment process is repeated.

[0014] A method for controlling temperature and humidity of a clinical drug refrigerator, comprising: Step 1: Collect data on temperature and humidity related parameters in the target refrigerator to obtain corresponding temperature and humidity data; Step 2: Analyze the collected temperature and humidity data, and determine whether the corresponding temperature and humidity data are abnormal based on the data analysis results. If abnormal, generate corresponding control instructions, which include temperature control instructions and humidity control instructions; Step 3: obtaining the required control power based on the obtained control instruction, and generating a corresponding control signal based thereon; Step 4: Adjust the temperature and humidity data of the corresponding refrigerator based on the obtained control signal, and provide information feedback based on the adjustment result.

[0015] A refrigerating cabinet comprises a main compressor and a sub compressor, a corresponding control terminal and a computer program stored in the control terminal and capable of realizing temperature and humidity data regulation.

[0016] The technical effects and advantages of a clinical medicine refrigerator temperature and humidity control system, method and refrigerator of the present invention are as follows: 1. Through multi-terminal data collection, reasonable node layout and accurate identification and analysis in the data collection module, temperature, humidity and pressure parameters are obtained in all directions and at multiple points, laying the foundation for accurate control; the data analysis module constructs a two-dimensional coordinate system, enthalpy curve and fluctuation coefficient comparison, deeply analyzes the data, accurately determines anomalies and generates instructions, and determines the control direction according to the trend of the local curve, ensuring that the temperature and humidity accurately meet the storage requirements of the drug, avoiding the risk of deterioration and extending the validity period; 2. For temperature control, an adaptive function is used in combination with monitoring data to optimize the proportional, integral, and differential factors to determine the power. The power is adjusted based on the adaptive function combined with the mean square error to determine the compensation coefficient, accurately matching the control needs, saving energy and ensuring the system is stable and efficient, reducing energy consumption, reducing losses, shortening response time, and improving equipment life and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a schematic diagram of a temperature and humidity control system for a clinical drug refrigerator of the present invention; Figure 2 It is a schematic diagram of a temperature and humidity control method of a clinical drug refrigerator of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] Example 1

[0020] See also Figure 1 As shown, this embodiment is a temperature and humidity control system for a clinical drug refrigerator, comprising: A data acquisition module is used to collect data on temperature and humidity related parameters in the target refrigerator to obtain corresponding temperature and humidity data; A data analysis module is used to analyze the collected temperature and humidity data, and determine whether the corresponding temperature and humidity data are abnormal based on the data analysis results. If abnormal, a corresponding control instruction is generated, and the control instruction includes a temperature control instruction and a humidity control instruction; A power acquisition module, which acquires the required control power based on the obtained control instruction and generates a corresponding control signal based thereon; The equipment control module is used to adjust the temperature and humidity data of the corresponding refrigerator based on the obtained control signal, and provide information feedback based on the adjustment result; The modules are connected to each other via wired and / or wireless means to achieve data transmission between modules.

[0021] It should be further explained that, in the specific implementation process, the process of collecting data on the temperature and humidity related parameters in the target refrigerator and obtaining the corresponding temperature and humidity data includes: The data collection node is equipped with several collection nodes and identification terminals. The collection node is composed of a pressure sensing terminal, a temperature sensing terminal, a humidity sensing terminal, an ultra-high frequency RFID, and a circularly polarized 9Dbi gain ultra-high frequency antenna. The collection nodes are deployed in the target refrigerator according to the pre-set deployment method. The deployment method refers to the deployment method set according to the internal structure design of the refrigerator, including 3 panels on each side and 2 panels on the back close to the top and bottom; After the deployment is completed, the temperature, humidity and pressure parameters at different points in the target refrigerator are collected based on the collection nodes and transmitted to the corresponding identification terminal. The identification terminal identifies and analyzes the received temperature and humidity parameters to obtain the corresponding temperature, humidity and water vapor pressure data.

[0022] It should be further explained that, in the specific implementation process, the process of analyzing the collected temperature and humidity data and determining whether the corresponding temperature and humidity data is abnormal based on the data analysis results includes: Construct a two-dimensional rectangular coordinate system with time as the horizontal axis and temperature and humidity data as the vertical axis, and map the collected temperature and humidity data into the corresponding two-dimensional rectangular coordinate system to obtain the corresponding temperature and humidity change curve; Set the monitoring period, obtain the temperature and humidity data in the current monitoring period based on the temperature and humidity change curve, and obtain the temperature and humidity enthalpy value corresponding to the target refrigerator in the current monitoring period based on it; the formula for obtaining the corresponding temperature and humidity enthalpy value is: ; In the formula, , and They represent the constant-pressure specific heat, saturation temperature and latent heat of vaporization of dry air respectively; they are all known thermodynamic parameters and can be obtained from relevant thermodynamics manuals; and Respectively represent the Temperature and humidity values ​​at a certain time; Statement Temperature, humidity and enthalpy at the time; Indicates the atmospheric pressure value; Then, based on the obtained temperature and humidity enthalpy values, the enthalpy value change curve corresponding to the corresponding monitoring period is obtained; Based on the enthalpy change curve, obtain the enthalpy fluctuation coefficient corresponding to the temperature and humidity enthalpy value in the current monitoring period ; In the formula, Indicates The temperature and humidity enthalpy value corresponding to the time, Indicates the total number of moments in the monitoring period; Obtain the temperature and humidity enthalpy values ​​corresponding to adjacent historical monitoring cycles under normal refrigeration conditions and record them as ; The obtained temperature and humidity enthalpy and Perform a comparison, and determine whether the corresponding temperature and humidity data are abnormal based on the comparison results; like , indicating that there is no abnormality in the temperature and humidity data fluctuations in the current monitoring period, no other operations are performed; if , indicating that the temperature and humidity data fluctuations are abnormal, the temperature and humidity data in the current monitoring period are marked as abnormal; It is the data threshold, which is determined according to actual needs; Obtain the corresponding temperature data curve based on the temperature and humidity data after abnormal marking; and perform curve analysis on the obtained temperature data curve to obtain the corresponding temperature state and temperature trend, the temperature state includes high temperature state, low temperature state and normal state, and the temperature trend includes constant temperature trend, heating trend and cooling trend; The build interval is A local window is formed, and the current moment is used as the right edge point of the corresponding local window, and the corresponding temperature data curve is intercepted to obtain the corresponding local curve, and the temperature state and temperature trend corresponding to the corresponding local curve are obtained; is a fixed constant; Furthermore, based on the temperature state and temperature trend corresponding to the local curve, it is determined whether the current temperature needs to be adjusted. If necessary, a corresponding temperature control instruction is generated. If not, a corresponding humidity control instruction is generated. For example, if the temperature trend is a rising trend and the temperature state is high temperature, temperature control is required. It should be further explained that, in a specific implementation process, the process of obtaining the required control power based on the obtained control instruction and generating a corresponding control signal based thereon includes: The control power includes temperature control power and humidity control power; the control signal includes temperature and humidity control signal and humidity control signal; When a temperature control instruction is identified, the corresponding temperature control power is obtained based on the temperature data corresponding to the current monitoring period and in combination with a pre-set adaptive temperature adjustment function, and a corresponding temperature control signal is generated based thereon; The formula for the adaptive temperature adjustment function is: P wd = K _p [e(t)+ 1 T_ I ∫ 0 t e( t )d t + T _D de(t) dt ] ; In the formula, , and They respectively represent the best proportional factor, the best integral time factor and the best differential time factor obtained in advance; Indicates the required temperature control power; ; represents the error parameter, and Respectively represent the collected temperature data and the expected temperature data; Represents a time index; The process of obtaining the optimal proportional factor, the optimal integral time factor, and the optimal differential time factor includes: Obtain the temperature data in the current monitoring period and input it into the pre-built parameter control model to obtain the corresponding model output results, which include the required optimal proportional factor, optimal integral time factor and optimal differential time factor; It should be further explained that, in the specific implementation process, the construction process of the parameter control model includes: The backbone network of the parameter control model is defined as an improved BP neural network. The improved BP neural network is used to learn the nonlinear mapping relationship between temperature data and corresponding control parameters. The basic structure of the improved BP neural network is an input layer, a hidden layer and an output layer. The control parameters include an optimal proportional factor, an optimal integral time factor and an optimal differential time factor. The input layer is used to receive input temperature data, obtain corresponding deviation temperature data based on the input temperature data, and construct a corresponding error image based on the deviation temperature data; For example: after receiving the corresponding temperature data, the input layer obtains the collection time corresponding to the corresponding temperature data and constructs a timestamp based on it; Obtaining the temperature state and temperature trend corresponding to the corresponding temperature data, and setting the expected temperature corresponding to the corresponding temperature data based thereon; further, obtaining the corresponding expected temperature and the input temperature data for deviation calculation to obtain corresponding deviation temperature data; Normalizing the corresponding deviation temperature data, and encoding the normalized deviation temperature data into polar coordinates; encoding the temperature value and timestamp in the corresponding deviation temperature data into polar angle and polar diameter respectively; Then, the encoded deviation temperature data is divided into several data subsets, and each of the obtained data subsets is reconstructed into a matrix and visualized in the form of a heat map to obtain a corresponding error image; The hidden layer is used to receive the error image output by the input layer, perform convolution operations on it, and learn the nonlinear relationship between input and output; The formula for the convolution operation is: ; In the formula, For the The hidden layer The output vector of each neuron node; Indicates The output vector of the previous hidden layer; Represents the index of the neuron node in the hidden layer; is the total number of neuron nodes in the corresponding hidden layer, represents the bias term; Indicates The hidden layer The neuron nodes in the previous layer The connection weights between neuron nodes; is a natural number, and For the The index and total number of neuron nodes in the previous hidden layer; represents the activation function; The output layer is used to receive the output vector of the hidden layer and map it into the required control parameters; Acquire temperature data corresponding to several historical monitoring cycles and their corresponding temperature control powers; and construct a corresponding training data set based thereon; the training data set consists of several training samples; Initialize the corresponding improved BP neural network, and input the corresponding training samples into the improved BP neural network after the initialization is completed. The improved BP neural network performs forward propagation operation with the input layer, hidden layer and output layer, and obtains the corresponding model loss error. The obtained model loss error is compared with the preset expected loss error. If the model loss error is less than the expected loss error, a new training sample is input to continue the forward propagation operation, and so on. If the model loss error is not less than the expected loss error, a back propagation operation is performed with the input layer, hidden layer and output layer, and the connection weights of the hidden layer and the output layer in the corresponding improved BP neural network are updated based on the preset learning algorithm. After the update is completed, repeat the corresponding forward propagation operation and back propagation operation process for iterative training, and record the corresponding model loss error until the obtained model loss error is less than the expected loss error and the model loss error of several consecutive training samples tends to converge, then the model training is completed; It should be further explained that, in the specific implementation process, the process of initializing the corresponding improved BP neural network includes: Obtain the total number of neuron nodes in the corresponding improved BP neural network, and based on it, obtain the network model parameters required for optimization, the network model parameters include network parameters such as connection weights of each neuron node, node thresholds, and bias items; and construct the corresponding fitness function based on the network model parameters; Then, an initial particle swarm is constructed based on the network model parameters to be optimized, and the corresponding particle swarm parameters are obtained. The particle swarm parameters include parameters such as learning factor and particle swarm size; Obtain the initial velocity and initial position of each particle in the corresponding initial particle group, calculate the fitness of each particle based on the pre-constructed fitness function, and obtain the global optimal solution in the corresponding initial particle group, which is recorded as ; Then, the corresponding initial particle group is randomly divided to obtain several particle groups; the particle with the best fitness in the corresponding particle group is obtained respectively. and the worst fitness particle ; The worst fitness particle in the corresponding particle group Perform local updates to obtain the corresponding worst fitness particle ; The particle with the worst fitness The particle with the worst fitness Compare, if the particle with the worst fitness Better than the worst fitness particle , then use the particle with the worst fitness Replace the original worst fitness particle ; If the worst fitness particle Not better than the worst fitness particle , then based on the global optimal solution Replace the best fitness particle , and perform local search again to obtain new particles with poor fitness , and compare it with the particle with the worst fitness Perform a second comparison, if it is still not better than the particle with the worst fitness , then a fitness particle is randomly generated to replace the original worst fitness particle And so on, until the number of local updates reaches Second-rate, is a fixed constant; After the local update is completed, the updated particle groups are summarized to obtain a new initial particle group, which is randomly grouped again, and the corresponding local update process is repeated to perform mixed iterative training on the corresponding initial particle group until it meets the preset stop condition. The stop condition refers to the corresponding mixed iterative training process reaching the preset maximum number of iterations; Then, based on the initial particle swarm after hybrid iterative training, the optimal network model parameters are obtained and applied to the corresponding improved BP neural network. Among them, the update formula of the particle initial velocity is: ; In the formula, and represents the learning factor, and is a random constant, and ; Represents particles The corresponding historical individual optimal value; represents the historical global optimal value; and Respectively represent Second and The initial velocity corresponding to the sub-particle iteration; Represents particles The position of the corresponding particle at the time of iteration; Represents the particle weight value; The update formula for the initial position of the particle is: ; The formula for local update is: ; , is a random constant; Indicates the maximum position distance deviation allowed for the corresponding particle; Indicates constraints; It should be further explained that, in the specific implementation process, the formula for updating the weight is: ; In the formula, and They represent the parameters controlling the model in the iterative training process. Second and The adjustment amount corresponding to the secondary connection weight; represents differential operation; and Represent the learning rate and smoothing factor respectively; Indicates Deviation temperature data during model iteration training; the learning algorithm is selected by those skilled in the art based on actual needs, and the present invention will not elaborate on it in detail; Represent the indexes of neuron nodes in the hidden layer and output layer respectively; When the identified control instruction is a humidity control instruction, an adjustment cycle is set, and humidity data within the corresponding adjustment cycle is obtained based on the temperature and humidity data curve, and the optimal humidity parameters corresponding to each moment in the corresponding adjustment cycle are obtained based on the pre-set humidity adaptive function. ; Among them, the humidity adaptive function is used to show the changing relationship between the corresponding humidity data and the temperature and humidity enthalpy value; The humidity adaptive function is defined as: ; In the formula, is the gas constant for dry air; is the Kelvin temperature corresponding to the current temperature data; is the temperature data at the current moment; Indicates the enthalpy deviation between the initial moment and the current moment in the adjustment period; represents the atmospheric pressure, Indicates the water vapor pressure data in the refrigerator at the current moment; Indicates the volume of the refrigerator; The mean square error values ​​of the corresponding humidity data and the optimal humidity parameters are respectively obtained, and the power compensation coefficient of the corresponding sub-compressor is obtained based thereon. Then, the corresponding first output power is adjusted based on the obtained power compensation coefficient to obtain the corresponding humidity control power, and a corresponding humidity control signal is generated based thereon; wherein, the first output power refers to the real-time output power or rated power of the corresponding sub-compressor; for example: if the sub-compressor is in working state, the first output power is the real-time output power, and if it is in shutdown state, the first output power is the rated power.

[0023] It should be further explained that, in the specific implementation process, the temperature and humidity data of the corresponding refrigerator are adjusted based on the obtained control signal, and the process of providing information feedback based on the adjustment result includes: Identify the received control signal; if the control signal is a temperature control signal, use the obtained temperature control power as the output power to control the operation of the main compressor in the refrigerator, adjust the temperature data of the corresponding refrigerator, and monitor the temperature in the corresponding refrigerator in real time; when the real-time temperature in the refrigerator is within the expected temperature range, obtain the corresponding real-time humidity control power based on the acquisition process of the humidity control power, and control the operation of the auxiliary compressor in the refrigerator based on it, and adjust the humidity data of the corresponding refrigerator synchronously until the real-time temperature in the refrigerator is within the expected temperature range and is in a constant temperature state; then construct multiple adjustment cycles to update the corresponding humidity control power until the corresponding real-time humidity value is within the expected humidity range and is the optimal humidity value corresponding to the current temperature; (that is, the temperature and humidity data in the refrigerator are in a dynamic equilibrium state within the expected temperature and humidity range); then stop the operation of the corresponding auxiliary compressor; wherein, the auxiliary compressor adopts a delayed shutdown strategy, that is, the auxiliary compressor will stop running only after the duration of the temperature and humidity data in the refrigerator being in a dynamic equilibrium state within the expected temperature and humidity range reaches a preset time threshold; If the control signal is a humidity control signal, repeat the above humidity data adjustment process; It should be further explained that, in the specific implementation process, the main compressor and the auxiliary compressor are structural equipment of the corresponding refrigerator itself, which are used to control the temperature and humidity of the refrigerator respectively; An embodiment of the present invention further includes: when the temperature data in the corresponding refrigerator is lower than a preset low temperature threshold, in order to prevent frost in the refrigerator, the refrigerator will adopt an evaporator air cooling defrosting mode; for example: automatically starting the circulation fan when the refrigeration temperature drops to a certain temperature, and turning off the fan when the temperature rises to a certain temperature.

[0024] The present invention acquires the temperature and humidity data in the refrigerator in real time and accurately through the data acquisition module, and performs in-depth analysis based on the acquired temperature and humidity data to accurately determine whether the temperature and humidity data are abnormal; it not only improves the judgment accuracy of the system, but also helps to timely discover and deal with potential temperature and humidity problems, ensuring that the medicines in the refrigerator are in the best storage environment; according to the results of the data analysis module, the corresponding control instructions are intelligently generated, and the acquisition of the required control power not only improves the control efficiency of the system, but also helps to achieve the rational use of energy and reduce energy consumption; based on the obtained control signal, the temperature and humidity data of the refrigerator are accurately adjusted, and through real-time monitoring and feedback mechanism, it is ensured that the temperature and humidity data in the refrigerator are always within the expected range; it not only improves the stability and reliability of the system, but also helps to ensure the quality and safety of the medicines.

[0025] Example 2

[0026] See also Figure 2 As shown, the part not described in detail in this embodiment is described in Example 1, which provides a method for controlling temperature and humidity of a clinical drug refrigerator, comprising: Step 1: Collect data on temperature and humidity related parameters in the target refrigerator to obtain corresponding temperature and humidity data; Step 2: Analyze the collected temperature and humidity data, and determine whether the corresponding temperature and humidity data are abnormal based on the data analysis results. If abnormal, generate corresponding control instructions, which include temperature control instructions and humidity control instructions; Step 3: obtaining the required control power based on the obtained control instruction, and generating a corresponding control signal based thereon; Step 4: Adjust the temperature and humidity data of the corresponding refrigerator based on the obtained control signal, and provide information feedback based on the adjustment result.

[0027] Example 3

[0028] The present embodiment discloses a refrigerator, including main and auxiliary compressors and corresponding control terminals, and a computer program stored on the control terminal and capable of realizing temperature and humidity data regulation. When the control terminal executes the computer program, the operation mode of the temperature and humidity control method of a clinical drug refrigerator provided above is realized.

[0029] Since the refrigerator described in this embodiment is a refrigerator used to implement a clinical drug refrigerator temperature and humidity control method in the embodiment of the present application, based on the clinical drug refrigerator temperature and humidity control method described in the embodiment of the present application, the technical personnel of the field can understand the specific implementation of the refrigerator of the present embodiment and its various variations, so how the refrigerator implements the method in the embodiment of the present application is not described in detail here. As long as the technical personnel of the field implement the refrigerator used in the clinical drug refrigerator temperature and humidity control method in the embodiment of the present application, it belongs to the scope of protection of the present application.

[0030] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0031] The above are only preferred embodiments of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technical users in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A temperature and humidity control system for a clinical drug refrigerator, characterized in that: include: A data acquisition module is used to collect data on temperature and humidity related parameters in the target refrigerator to obtain corresponding temperature and humidity data; A data analysis module is used to analyze the collected temperature and humidity data, and determine whether the corresponding temperature and humidity data are abnormal based on the data analysis results. If abnormal, a corresponding control instruction is generated, and the control instruction includes a temperature control instruction and a humidity control instruction; A power acquisition module, which acquires the required control power based on the obtained control instruction and generates a corresponding control signal based thereon; The equipment control module adjusts the temperature and humidity data of the corresponding refrigerator based on the obtained control signal, and provides information feedback based on the adjustment result.

2. The temperature and humidity control system for clinical drug refrigerators according to claim 1 is characterized in that: The process of collecting data on temperature and humidity related parameters in the target refrigerator and obtaining corresponding temperature and humidity data includes: The data collection node is provided with a plurality of collection nodes and identification terminals; based on the collection nodes, data is collected on the temperature, humidity and pressure parameters at different points in the target refrigerator, and the data are transmitted to the corresponding identification terminal, and the identification terminal performs parameter identification and analysis on the received temperature and humidity parameters to obtain corresponding temperature and humidity data and water vapor pressure data.

3. The temperature and humidity control system for clinical drug refrigerators according to claim 1 is characterized in that: The process of analyzing the collected temperature and humidity data and determining whether the corresponding temperature and humidity data are abnormal based on the data analysis results includes: Construct a temperature and humidity change curve; and set a monitoring period, obtain the temperature and humidity data in the current monitoring period based on the temperature and humidity change curve, and obtain the temperature and humidity enthalpy value corresponding to the target refrigerator in the current monitoring period based on the temperature and humidity change curve; Based on the obtained temperature and humidity enthalpy values, an enthalpy value change curve corresponding to the corresponding monitoring period is obtained; Based on the enthalpy change curve, the enthalpy fluctuation coefficient corresponding to the temperature and humidity enthalpy value in the current monitoring period is obtained. ; Obtain the temperature and humidity enthalpy values ​​corresponding to adjacent historical monitoring cycles under normal refrigeration conditions and record them as ; The obtained temperature and humidity enthalpy and Perform a comparison, and determine whether the corresponding temperature and humidity data are abnormal based on the comparison results; like , no other operations are performed; if , the temperature and humidity data in the current monitoring period are marked as abnormal; is the data threshold; Obtain a corresponding temperature data curve based on the temperature and humidity data after abnormal marking; and perform curve analysis on the obtained temperature data curve to obtain a corresponding temperature state and temperature trend, wherein the temperature state includes a high temperature state, a low temperature state and a normal state, and the temperature trend includes a constant temperature trend, a temperature increase trend and a temperature decrease trend; The build interval is A local window is selected, and the current moment is used as the right edge point of the corresponding local window, and the corresponding temperature data curve is intercepted to obtain the corresponding local curve, and the temperature state and temperature trend corresponding to the corresponding local curve are obtained; is a fixed constant; Based on the temperature state and temperature trend corresponding to the local curve, it is judged whether the current temperature needs to be adjusted. If necessary, a corresponding temperature control instruction is generated. If not, a corresponding humidity control instruction is generated.

4. The temperature and humidity control system for clinical drug refrigerators according to claim 3 is characterized in that: The process of obtaining the required control power based on the obtained control instruction and generating a corresponding control signal based on the control instruction includes: The control power includes temperature control power and humidity control power; the control signal includes temperature and humidity control signal and humidity control signal; Identify the obtained control instruction, and when the temperature control instruction is identified, obtain the corresponding temperature control power based on the temperature data corresponding to the current monitoring period and in combination with the preset adaptive temperature adjustment function, and generate the corresponding temperature control signal based on the temperature data; When the identified control instruction is a humidity control instruction, an adjustment cycle is set, and humidity data within the corresponding adjustment cycle is obtained based on the temperature and humidity data curve, and the optimal humidity parameters corresponding to each moment in the corresponding adjustment cycle are obtained based on the pre-set humidity adaptive function. ; The mean square error values ​​of the corresponding humidity data and the optimal humidity parameters are respectively obtained, and the power compensation coefficient of the corresponding auxiliary compressor is obtained based thereon, the corresponding first output power is adjusted based on the obtained power compensation coefficient, the corresponding humidity control power is obtained, and the corresponding humidity control signal is generated based thereon; wherein the first output power is the real-time output power or rated power of the auxiliary compressor in the refrigerator.

5. The temperature and humidity control system for clinical drug refrigerators according to claim 4 is characterized in that: The formula of the adaptive temperature adjustment function is: ; In the formula, , and They respectively represent the best proportional factor, the best integral time factor and the best differential time factor obtained in advance; Indicates the required temperature control power; ; represents the temperature error parameter, and Respectively represent the collected temperature data and the expected temperature data; Represents a time index; The humidity adaptive function is defined as: ; In the formula, is the gas constant for dry air; is the Kelvin temperature corresponding to the current temperature data; is the temperature data at the current moment; Indicates the temperature and humidity enthalpy deviation between the initial moment and the current moment in the adjustment cycle; represents the atmospheric pressure, Indicates the water vapor pressure data in the refrigerator at the current moment; Indicates the volume of the refrigerator.

6. The temperature and humidity control system for clinical drug refrigerators according to claim 4, characterized in that: The process of obtaining the optimal proportional factor, the optimal integral time factor, and the optimal differential time factor includes: Obtain the temperature data in the current monitoring period and input it into the pre-built parameter control model to obtain the corresponding model output results, which include the required optimal proportional factor, optimal integral time factor and optimal differential time factor; The construction process of the corresponding parameter control model is: The backbone network of the parameter control model is defined as an improved BP neural network, wherein the basic structure of the improved BP neural network is an input layer, a hidden layer and an output layer; The input layer is used to receive input temperature data, obtain corresponding deviation temperature data based on the input temperature data, and construct a corresponding error image based on the deviation temperature data; The hidden layer is used to receive the error image output by the input layer, perform convolution operation on it, and learn the nonlinear relationship between input and output; The formula for the convolution operation is: ; In the formula, For the The hidden layer The output vector of each neuron node; Indicates The output vector of the previous hidden layer; Represents the index of the neuron node in the hidden layer; is the total number of neuron nodes in the corresponding hidden layer, represents the bias term; Indicates The hidden layer The neuron nodes in the previous layer The connection weights between neuron nodes; is a natural number, and For the The index and total number of neuron nodes in the previous hidden layer; represents the activation function; The output layer is used to receive the output vector of the hidden layer and map it into required control parameters, wherein the control parameters include an optimal proportional factor, an optimal integral time factor, and an optimal differential time factor; Acquire temperature data corresponding to a number of historical monitoring cycles and their corresponding temperature control powers; and construct a corresponding training data set based thereon; the training data set is composed of a number of training samples; Initialize the corresponding improved BP neural network, and input the corresponding training samples into the improved BP neural network after the initialization, the improved BP neural network performs forward propagation operation with input layer, hidden layer and output layer, and obtains the corresponding model loss error, compares the obtained model loss error with the preset expected loss error, if the model loss error is less than the expected loss error, then input a new training sample to continue the forward propagation operation, and so on; if the model loss error is not less than the expected loss error, then perform back propagation operation with input layer, hidden layer and output layer as the order, and update the connection weights of the hidden layer and the output layer in the corresponding improved BP neural network based on the preset learning algorithm; The formula for updating weights is: ; In the formula, and They represent the parameters controlling the model in the iterative training process. Second and The adjustment amount corresponding to the secondary connection weight; represents differential operation; and Represent the learning rate and smoothing factor respectively; ) indicates the Deviation temperature data during model iteration training; After the update is completed, repeat the corresponding forward propagation and back propagation operations for iterative training, and record the corresponding model loss error until the obtained model loss error is less than the expected loss error and the model loss error of several consecutive training samples tends to converge, then the model training is completed.

7. The temperature and humidity control system for clinical drug refrigerators according to claim 6, characterized in that: The process of initializing the corresponding improved BP neural network includes: Obtaining the total number of neuron nodes in the corresponding improved BP neural network, and obtaining the network model parameters required for optimization based on the total number of neuron nodes; and constructing the corresponding fitness function based on the network model parameters; Construct an initial particle swarm based on the network model parameters to be optimized, and obtain the corresponding particle swarm parameters; Obtain the initial velocity and initial position of each particle in the corresponding initial particle group, calculate the fitness of each particle based on the pre-constructed fitness function, and obtain the global optimal solution in the corresponding initial particle group, which is recorded as ; The corresponding initial particle group is randomly divided to obtain several particle groups; the optimal fitness particles in the corresponding particle groups are obtained respectively. and the worst fitness particle ; The worst fitness particle in the corresponding particle group Perform local updates to obtain the corresponding worst fitness particle ; The particle with the worst fitness The particle with the worst fitness Compare, if the particle with the worst fitness Better than the worst fitness particle , then use the particle with the worst fitness Replace the original worst fitness particle ; If the worst fitness particle Not better than the worst fitness particle , then based on the global optimal solution Replace the best fitness particle , and perform local search again to obtain new particles with poor fitness , and compare it with the particle with the worst fitness Perform a second comparison, if it is still not better than the particle with the worst fitness , then a fitness particle is randomly generated to replace the original worst fitness particle And so on, until the number of local updates reaches L times, where L is a fixed constant; After the local update is completed, the updated particle groups are summarized to obtain a new initial particle group, which is randomly grouped again, and the corresponding local update process is repeated to perform mixed iterative training on the corresponding initial particle group until it meets the preset stop condition; Based on the initial particle swarm after hybrid iterative training, the required optimal network model parameters are obtained and applied to the corresponding improved BP neural network.

8. The temperature and humidity control system for clinical drug refrigerators according to claim 7, characterized in that: The process of adjusting the temperature and humidity data of the corresponding refrigerator based on the obtained control signal and providing information feedback based on the adjustment result includes: Identify the received control signal; if the control signal is a temperature control signal, use the obtained temperature control power as the output power to control the operation of the main compressor in the refrigerator, and adjust the temperature data of the corresponding refrigerator. At the same time, based on the acquisition process of the humidity control power, obtain the required real-time humidity control power, and control the operation of the auxiliary compressor in the refrigerator based on it, and synchronously adjust the humidity data of the corresponding refrigerator until the real-time temperature in the refrigerator is within the expected temperature range and the temperature data is in a constant temperature state; When the temperature data is in a constant temperature state, multiple adjustment cycles are constructed to update the corresponding humidity control power until the corresponding real-time humidity value is within the expected humidity range and is the optimal humidity value corresponding to the current temperature; then the operation of the corresponding auxiliary compressor is stopped; If the control signal is a humidity control signal, the above humidity data adjustment process is repeated.

9. A method for controlling temperature and humidity of a clinical medicine refrigerator, applied to the temperature and humidity control system of a clinical medicine refrigerator as claimed in any one of claims 1 to 8, characterized in that: include: Step 1: Collect data on temperature and humidity related parameters in the target refrigerator to obtain corresponding temperature and humidity data; Step 2: Analyze the collected temperature and humidity data, and determine whether the corresponding temperature and humidity data are abnormal based on the data analysis results. If abnormal, generate corresponding control instructions, which include temperature control instructions and humidity control instructions; Step 3: obtaining the required control power based on the obtained control instruction, and generating a corresponding control signal based thereon; Step 4: Adjust the temperature and humidity data of the corresponding refrigerator based on the obtained control signal, and provide information feedback based on the adjustment result.

10. A refrigerator, comprising a main compressor and a secondary compressor and a corresponding control terminal and a computer program stored in the control terminal and capable of controlling temperature and humidity data, characterized in that: When the control terminal executes the computer program, the temperature and humidity control method for the clinical drug refrigerator according to claim 9 is implemented.

Images