A biochemical material storage system based on the Internet of Things

Abstract

The present invention discloses a biochemical material storage system based on the Internet of Things, comprising: a real-time monitoring module: using sensors to collect storage environment data and quality change data of biochemical materials in real time, relating to the technical field of biochemical material storage, the setting of a storage environment condition scheme generation module, taking a quality change standard function as a benchmark, combining historical data with quality change prediction and energy consumption prediction models to screen an initial group, and iteratively optimizing through a genetic algorithm. During the iterative process, on the one hand, high-quality individuals are screened for cross-mutation with the lowest energy consumption as the guide, and the energy consumption optimal solution is continuously approached; on the other hand, based on dynamic setting conditions, individuals with relatively low energy consumption and better quality change are promptly incorporated into the optimization process, ensuring that the scheme is both energy-saving and conducive to the preservation of biochemical materials, and customizing storage conditions for the storage of biochemical materials that are both conducive to preservation and low in energy consumption.

Description

Technical Field

[0001] The present invention relates to the technical field of biochemical material storage, and in particular to a biochemical material storage system based on the Internet of Things. Background Art

[0002] Biochemical materials refer to the various raw materials, intermediates, and products used in the biochemical industry. Within this sector, material storage has always been a critical factor affecting product quality, production costs, and the sustainable development of the enterprise. With the continuous advancement of technology and the continued expansion of the industry, traditional material storage methods have increasingly exposed numerous limitations, making them unable to meet the current demands of refined, intelligent, and green industrial development.

[0003] Current storage of biochemical materials presents challenges. These materials often have complex compositions and are highly sensitive to the storage environment. Slight fluctuations in temperature, slight changes in humidity, altered gas composition, and even variations in storage pressure can trigger chemical reactions in the materials, leading to quality degradation, such as degradation of active ingredients, increased impurities, and loss of activity. These quality degradations not only directly impact the performance and stability of subsequent products, resulting in lower yields and higher rates of defective and scrapped products, but can also lead to significant material losses and market reputational risks for companies. In the past, due to a lack of accurate and real-time monitoring methods, companies often struggled to detect gradual quality changes during storage, relying instead on periodic spot checks. This belated approach, tantamount to "fixing the fold after the horse has bolted," failed to fundamentally guarantee material quality. Furthermore, to maintain a suitable storage environment, various environmental control devices, such as air conditioners, dehumidifiers, and ventilation systems, are widely used. However, this also results in significant energy consumption. Traditional environmental control systems often rely on fixed parameters based on empirical experience, failing to fully consider the real-time state of the materials, dynamic environmental changes, and the optimal balance of energy consumption.

[0004] This results in actual operations, where either the pursuit of stable material quality leads to blindly increasing environmental control intensity, resulting in a large amount of energy waste; or excessive focus on energy consumption control leads to ignoring the potential harm of environmental fluctuations to material quality, ultimately outweighing the benefits. In order to obtain a storage environment condition solution that is both conducive to material preservation and achieves low energy consumption, this application proposes a biochemical material storage system based on the Internet of Things. Summary of the Invention

[0005] In order to solve the technical problems in actual operation, either the excessive pursuit of stable material quality leads to blindly increasing the intensity of environmental control, resulting in a large amount of energy waste; or the excessive focus on energy consumption control leads to neglecting the potential harm of environmental fluctuations to material quality, which ultimately outweighs the gains. The present invention proposes a biochemical material storage system based on the Internet of Things, with the aim of obtaining a storage environment condition solution that is both conducive to material preservation and can achieve low energy consumption.

[0006] The present invention proposes a biochemical material storage system based on the Internet of Things, comprising:

[0007] Real-time monitoring module: Through sensors, it collects storage environment data and material quality change data of biochemical materials in real time, and obtains energy consumption data, and transmits the storage environment data, material quality change data, and energy consumption data to the data storage module in real time;

[0008] Data storage module: stores data from the real-time monitoring module;

[0009] Quality change prediction and energy consumption prediction model generation module: Analyzes the storage environment data, material quality change data, and energy consumption data of the data storage module to build quality change prediction and energy consumption prediction models;

[0010] Storage environment condition program generation module: For a certain biochemical material, a quality change standard function is set. According to historical environmental parameters, through quality change prediction and energy consumption prediction models, multiple sets of environmental parameters lower than the quality change standard function are obtained. Then, through data analysis, with the minimum energy consumption as the goal, a set of target environmental parameters is obtained.

[0011] Preferably, the sensors include temperature sensors, humidity sensors, pressure sensors, gas sensors, material weight sensors, and liquid level sensors.

[0012] Preferably, the quality change prediction and energy consumption prediction models are constructed as follows:

[0013] The storage environment data and material quality change data of the data storage module are analyzed. The quality change curve of a biochemical material under different environmental conditions per unit time is calculated using the support vector regression algorithm. A quality change curve prediction model is constructed. In the quality change curve prediction model, the environmental parameters are used as input, and the quality change curve prediction result per unit time is used as output.

[0014] The storage environment data and energy consumption data of the data storage module are analyzed. The energy consumption of a certain biochemical material under different environmental conditions per unit time is calculated using the support vector regression algorithm. An energy consumption prediction model is constructed. In the energy consumption prediction model, environmental parameters are used as input, and the energy consumption prediction result per unit time is used as output.

[0015] The quality change curve prediction model and the energy consumption prediction model are integrated to construct quality change prediction and energy consumption prediction models. When new environmental parameters are input, they are input into the quality change curve prediction model and the energy consumption prediction model respectively. The quality change curve prediction model outputs the prediction result of the quality change curve per unit time, and the energy consumption prediction model outputs the energy consumption prediction result per unit time.

[0016] Preferably, in the storage environment condition scheme generation module, the target environment parameters are obtained by the following method:

[0017] S1. For a certain biochemical material, set a quality change standard function. According to the historical environmental parameters, through the quality change prediction and energy consumption prediction model, obtain multiple groups of environmental parameters lower than the quality change standard function, and set them as the initial group. ;

[0018] S2. For each environmental parameter that is lower than the quality change standard function, calculate the corresponding energy consumption through the quality change prediction and energy consumption prediction model, and select the X environmental parameters with the lowest energy consumption;

[0019] S3, X environmental parameters are subjected to crossover and mutation operations through genetic algorithms to obtain multiple new environmental parameters, and the quality change curve of each environmental parameter is calculated to obtain multiple sets of environmental parameters that are lower than the quality change standard function, and the obtained multiple sets of environmental parameters that are lower than the quality change standard function are compared with the Merge to form new groups ;

[0020] S4. Repeat steps S2 and S3. After multiple rounds of iterations, when the absolute value of the difference in the minimum energy consumption obtained in Y consecutive iterations is less than the set threshold, the environmental parameter with the lowest energy consumption is taken as the target environmental parameter.

[0021] Preferably, X is a non-negative even number, and 4≤X;

[0022] Y is a positive integer, and 3≤X.

[0023] Preferably, assume that there are n iterations, for the i-th iteration, i=1,2,...,n, X environmental parameters obtain X energy consumption values, X quality change curves, and the energy consumption value with the largest of the X energy consumption values ​​is set as E;

[0024] For the population of the i-th iteration ,group In the process, if there is an environmental parameter whose energy consumption is not greater than 1.02E and whose quality change curve is lower than any of the X quality change curves, it will be brought into the crossover and mutation operation of the X environmental parameters.

[0025] Preferably, during the iteration process, the energy consumption values ​​and quality change curve values ​​corresponding to the environmental parameters in the current group are visualized in real time to form a two-dimensional scatter plot or a Pareto frontier diagram;

[0026] The quality change standard function and genetic algorithm parameters can be manually adjusted based on the visualization results.

[0027] A biochemical material storage method based on the Internet of Things comprises the following steps:

[0028] Through sensors, the storage environment data and material quality change data of biochemical materials are collected in real time, and energy consumption data is obtained, and the storage environment data, material quality change data, and energy consumption data are transmitted to the data storage module in real time;

[0029] Extract storage environment data, material quality change data and energy consumption data from the data storage module, use support vector regression algorithm to build a quality change curve prediction model and an energy consumption prediction model, integrate the quality change curve prediction model and the energy consumption prediction model to form a quality change prediction and energy consumption prediction model, and after receiving the environmental parameter input, the quality change prediction and energy consumption prediction models respectively output the quality change curve prediction result and energy consumption prediction result per unit time;

[0030] For a certain biochemical material, a quality change standard function is set. Based on historical environmental parameters, multiple groups of environmental parameters that are lower than the quality change standard function are obtained through quality change prediction and energy consumption prediction models. These groups are used as the initial population. The energy consumption of each environmental parameter in the initial population is calculated through the quality change prediction and energy consumption prediction models. X environmental parameters with the lowest energy consumption are selected. These X parameters are cross-mutated and mutated using a genetic algorithm to generate new parameters. The quality change curve of the new parameters is then calculated using the quality change prediction and energy consumption prediction models. The remaining environmental parameters that are lower than the quality change standard function are merged with the initial population to form a new population, and multiple rounds of iterations are performed:

[0031] In the i-th iteration, i = 1, 2, ..., n, for X environmental parameters, obtain the corresponding X energy consumption values ​​and X quality change curves, find the maximum energy consumption value and record it as E, for the environmental parameter in the iteration group, if its energy consumption is not greater than 1.02E and its quality change curve is lower than any of the X quality change curves, then bring it into the crossover and mutation operation of the X environmental parameters;

[0032] Continue iterating until the absolute value of the difference in the lowest energy consumption obtained in Y consecutive iterations is less than the set threshold. The goal is to minimize energy consumption, and finally select the environmental parameter with the lowest energy consumption as the target environmental parameter.

[0033] During the iteration process, the energy consumption values ​​and quality change curve values ​​corresponding to the environmental parameters in the current group are visualized in real time, and the relationship between them is displayed through a two-dimensional scatter plot or Pareto frontier diagram;

[0034] Based on the visualization results, the parameters of the quality change standard function and the genetic algorithm can be manually adjusted to optimize the generation process of the storage environment condition scheme.

[0035] According to the final target environmental parameters, the storage environment of biochemical materials is regulated and controlled, and the operating parameters of relevant environmental control equipment (such as ventilation systems, dehumidifiers, air conditioners, heating devices, etc.) are adjusted to ensure that the materials are stored in an environment that meets quality requirements and has the lowest energy consumption.

[0036] The biochemical material storage system based on the Internet of Things proposed in the present invention has the following beneficial technical effects:

[0037] 1. The setting of the storage environment condition scheme generation module takes the quality change standard function as the benchmark, combines historical data with quality change prediction and energy consumption prediction models to screen the initial population, and uses genetic algorithm for iterative optimization. During the iterative process, on the one hand, high-quality individuals are selected for cross-mutation with the lowest energy consumption as the guide, and the optimal energy consumption solution is continuously approached; on the other hand, based on the dynamic setting conditions, individuals with relatively low energy consumption and better quality change are promptly included in the optimization process to ensure that the scheme is both energy-saving and conducive to the preservation of biochemical materials. This process is repeated many times until the target environmental parameters that meet the requirements of continuous, stable and low energy consumption are found, and storage conditions that are both conducive to preservation and low energy consumption are customized for the storage of biochemical materials.

[0038] 2. During the iteration process of generating storage environment condition solutions, the energy consumption values ​​and quality change curve values ​​corresponding to the environmental parameters in the current group are visualized in real time, forming a two-dimensional scatter plot or Pareto frontier diagram. This helps users intuitively observe the algorithm's search progress, understand the trade-off between energy consumption and quality, and promptly detect any algorithm anomalies, such as falling into a local optimum that causes points to cluster in a certain area. Based on the visualization results, the quality change standard function and genetic algorithm parameters can be manually adjusted to achieve human-machine collaborative optimization. This makes the generation of storage environment solutions no longer a mechanical algorithm execution, but a flexible customization that incorporates human experience and wisdom. This further improves the accuracy and practicality of the solution and increases satisfaction with the final target environmental parameters.

[0039] 3. By collecting biochemical material storage environment data and material quality change data in real time, and obtaining energy consumption data at the same time, this real-time data collection can effectively avoid material quality problems and energy waste caused by information lag.

[0040] 4. The quality change prediction and energy consumption prediction model generation modules are set up. The quality change prediction model utilizes a support vector regression algorithm to deeply analyze the inherent correlation between environmental parameters and material quality change curves. By inputting environmental parameters, it accurately outputs material quality change trends per unit time. This enables companies to proactively assess material stability under different storage conditions, plan material turnover and usage in advance, minimize the risk of material deterioration, ensure consistent product quality, reduce defective product output, and achieve significant cost savings. The energy consumption prediction model also utilizes a support vector regression algorithm to clearly reveal the quantitative relationship between environmental factors and energy consumption. During production operations, companies can use this model to proactively estimate energy consumption under different environmental control strategies, assisting in developing scientifically sound energy-saving plans, optimizing energy allocation, reducing operating costs, and improving economic efficiency, in line with the current green and low-carbon development philosophy. The integrated quality change and energy consumption prediction model allows for a single input of environmental parameters and simultaneous output of both key information on quality change and energy consumption. This provides one-stop support for comprehensive decision-making, avoids the tedious switching and comparison between different models, and significantly improves decision-making efficiency.

[0041] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a principle block diagram of the system of the present invention;

[0043] Figure 2 Flowchart of the method of the present invention. DETAILED DESCRIPTION

[0044] The following describes embodiments of the present invention in detail. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar symbols throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended only to explain the present invention, and are not to be construed as limiting the present invention.

[0045] like Figure 1 The biochemical material storage system based on the Internet of Things shown includes:

[0046] Real-time monitoring module: Through sensors, it collects storage environment data and material quality change data of biochemical materials in real time, and obtains energy consumption data, and transmits the storage environment data, material quality change data, and energy consumption data to the data storage module in real time;

[0047] Sensors include temperature sensors, humidity sensors, pressure sensors, gas sensors (capable of detecting harmful gas concentrations), material weight sensors, and liquid level sensors;

[0048] By collecting biochemical material storage environment data and material quality change data in real time, as well as energy consumption data, this real-time data collection can effectively avoid material quality problems and energy waste caused by information lag.

[0049] Data storage module: stores data from the real-time monitoring module;

[0050] Quality change prediction and energy consumption prediction model generation module: Analyzes the storage environment data, material quality change data, and energy consumption data of the data storage module to build quality change prediction and energy consumption prediction models;

[0051] Construct quality change prediction and energy consumption prediction models as follows:

[0052] The storage environment data and material quality change data of the data storage module are analyzed. The quality change curve of a biochemical material under different environmental conditions per unit time is calculated using the support vector regression algorithm. A quality change curve prediction model is constructed. In the quality change curve prediction model, the environmental parameters are used as input, and the quality change curve prediction result per unit time is used as output.

[0053] The storage environment data and energy consumption data of the data storage module are analyzed. The energy consumption of a certain biochemical material under different environmental conditions per unit time is calculated using the support vector regression algorithm. An energy consumption prediction model is constructed. In the energy consumption prediction model, environmental parameters are used as input, and the energy consumption prediction result per unit time is used as output.

[0054] The quality change curve prediction model and the energy consumption prediction model are integrated to construct quality change prediction and energy consumption prediction models. When new environmental parameters are input, they are input into the quality change curve prediction model and the energy consumption prediction model respectively. The quality change curve prediction model outputs the prediction result of the quality change curve per unit time, and the energy consumption prediction model outputs the prediction result of energy consumption per unit time.

[0055] The quality change prediction and energy consumption prediction model generation modules are set up. The quality change prediction model utilizes a support vector regression algorithm to deeply analyze the inherent correlation between environmental parameters and material quality change curves. By inputting environmental parameters, it accurately outputs material quality change trends per unit time. This enables companies to proactively assess material stability under different storage conditions, plan material turnover and usage in advance, minimize the risk of material deterioration, ensure product quality consistency, reduce defective product output, and achieve significant cost savings. The energy consumption prediction model also utilizes a support vector regression algorithm to clearly reveal the quantitative relationship between environmental factors and energy consumption. During production operations, companies can use this model to proactively estimate energy consumption under different environmental control strategies, assisting in developing scientifically sound energy-saving plans, optimizing energy allocation, reducing operating costs, and improving economic efficiency, all in line with the current green and low-carbon development philosophy. The integrated quality change and energy consumption prediction model allows for a single input of environmental parameters and simultaneous output of both key information on quality change and energy consumption. This provides one-stop support for comprehensive decision-making, eliminating the tedious switching and comparison between different models and significantly improving decision-making efficiency.

[0056] Storage environment condition plan generation module: For a certain biochemical material, a quality change standard function is set. Based on historical environmental parameters, multiple sets of environmental parameters lower than the quality change standard function are obtained through quality change prediction and energy consumption prediction models. Then, through data analysis, a set of target environmental parameters is obtained with the minimum energy consumption as the goal.

[0057] In the storage environment condition scheme generation module, the target environment parameters are obtained through the following methods:

[0058] S1. For a certain biochemical material, set a quality change standard function. According to the historical environmental parameters, through the quality change prediction and energy consumption prediction model, obtain multiple groups of environmental parameters lower than the quality change standard function, and set them as the initial group. ;

[0059] An environmental parameter that is lower than the standard function of quality change means that the quality change curve of the environmental parameter is lower than the standard function of quality change;

[0060] S2. For each environmental parameter that is lower than the quality change standard function, calculate the corresponding energy consumption through the quality change prediction and energy consumption prediction model, and select the X environmental parameters with the lowest energy consumption;

[0061] In an optional embodiment, X is a non-negative even number, and 4≤X;

[0062] S3, X environmental parameters are subjected to crossover and mutation operations through genetic algorithms to obtain multiple new environmental parameters, and the quality change curve of each environmental parameter is calculated to obtain multiple sets of environmental parameters that are lower than the quality change standard function, and the obtained multiple sets of environmental parameters that are lower than the quality change standard function are compared with the Merge to form new groups ;

[0063] S4. Repeat steps S2 and S3. After multiple rounds of iterations, when the absolute value of the difference in the minimum energy consumption obtained in Y consecutive iterations is less than the set threshold, the environmental parameter with the lowest energy consumption is taken as the target environmental parameter.

[0064] In an optional embodiment, Y is a positive integer, and 3≤X;

[0065] Repeat steps S2 and S3 for multiple rounds of iterations, e.g. Repeat steps S2 and S3 to form a new group ,Then Repeat steps S2 and S3 to form a new group , thus forming a cycle, so that the group continues to evolve towards lower energy consumption while satisfying the standard function of quality change.

[0066] Assume there are n iterations. For the i-th iteration, i=1,2,...,n, X environmental parameters yield X energy consumption values ​​and X quality change curves. Let the energy consumption value with the largest X energy consumption values ​​be E.

[0067] For the population of the i-th iteration , group = In the , if there is an environmental parameter whose energy consumption is not greater than 1.02E and whose quality change curve is lower than any of the X quality change curves, then it will be brought into the crossover and mutation operations of the X environmental parameters, that is, the environmental parameters for the crossover and mutation operations through the genetic algorithm are no longer X, but the environmental parameters with energy consumption not greater than 1.02E and whose quality change curve is lower than any of the X quality change curves must be added.

[0068] The setting of the storage environment condition scheme generation module is based on the quality change standard function, and combines historical data with quality change prediction and energy consumption prediction models to screen the initial population, and iteratively optimizes through genetic algorithms. During the iterative process, on the one hand, high-quality individuals are selected for cross-mutation with the lowest energy consumption as the guide, and the optimal energy consumption solution is continuously approached; on the other hand, based on the dynamic setting conditions, individuals with relatively low energy consumption and better quality change are promptly included in the optimization process to ensure that the scheme is both energy-saving and conducive to the preservation of biochemical materials. This process is repeated for many rounds until the target environmental parameters that meet the requirements of continuous, stable and low energy consumption are found, and storage conditions that are conducive to preservation and low energy consumption are customized for the storage of biochemical materials.

[0069] During the iteration process, the energy consumption values ​​and quality change curve values ​​corresponding to the environmental parameters in the current group are visualized in real time to form a two-dimensional scatter plot or Pareto frontier diagram; this helps users intuitively observe the algorithm's search process, understand the trade-off between energy consumption and quality, and promptly detect whether the algorithm has abnormalities (such as falling into a local optimum causing points to cluster in a certain area). Based on the visualization results, the quality change standard function and genetic algorithm parameters can be manually adjusted to achieve human-computer collaborative optimization and improve the satisfaction with the final target environmental parameters.

[0070] Biochemical materials include:

[0071] Biomass raw materials: sugars, starch, cellulose, oils;

[0072] Microorganisms: bacteria, yeast, mold;

[0073] Bioactive substances and intermediates: enzymes, amino acids, organic acids;

[0074] Culture medium components: carbon source, nitrogen source, inorganic salts and trace elements.

[0075] This system aims to use Internet of Things technology to monitor the storage environment and material quality changes of biochemical materials in real time, ensure the quality and safety of materials, minimize energy consumption while ensuring material quality, reduce material loss and defective rate from the perspective of material quality assurance, ensure stable product quality and enhance market competitiveness; from the perspective of energy management, energy-saving regulation and control can reduce energy consumption expenses; and improve the efficiency and economy of storage management.

[0076] like Figure 2 The biochemical material storage method based on the Internet of Things shown includes the following steps:

[0077] Through sensors, the storage environment data and material quality change data of biochemical materials are collected in real time, and energy consumption data is obtained, and the storage environment data, material quality change data, and energy consumption data are transmitted to the data storage module in real time;

[0078] Extract storage environment data, material quality change data and energy consumption data from the data storage module, use support vector regression algorithm to build a quality change curve prediction model and an energy consumption prediction model, integrate the quality change curve prediction model and the energy consumption prediction model to form a quality change prediction and energy consumption prediction model, and after receiving the environmental parameter input, the quality change prediction and energy consumption prediction models respectively output the quality change curve prediction result and energy consumption prediction result per unit time;

[0079] For a certain biochemical material, a quality change standard function is set. Based on historical environmental parameters, multiple groups of environmental parameters that are lower than the quality change standard function are obtained through quality change prediction and energy consumption prediction models. These groups are used as the initial population. The energy consumption of each environmental parameter in the initial population is calculated through the quality change prediction and energy consumption prediction models. X environmental parameters with the lowest energy consumption are selected. These X parameters are cross-mutated and mutated using a genetic algorithm to generate new parameters. The quality change curve of the new parameters is then calculated using the quality change prediction and energy consumption prediction models. The remaining environmental parameters that are lower than the quality change standard function are merged with the initial population to form a new population, and multiple rounds of iterations are performed:

[0080] In the i-th iteration, i = 1, 2, ..., n, for X environmental parameters, obtain the corresponding X energy consumption values ​​and X quality change curves, find the maximum energy consumption value and record it as E, for the environmental parameter in the iteration group, if its energy consumption is not greater than 1.02E and its quality change curve is lower than any of the X quality change curves, then bring it into the crossover and mutation operation of the X environmental parameters;

[0081] Continue iterating until the absolute value of the difference in the lowest energy consumption obtained in Y consecutive iterations is less than the set threshold. The goal is to minimize energy consumption, and finally select the environmental parameter with the lowest energy consumption as the target environmental parameter.

[0082] During the iteration process, the energy consumption values ​​and quality change curve values ​​corresponding to the environmental parameters in the current group are visualized in real time, and the relationship between them is displayed through a two-dimensional scatter plot or Pareto frontier diagram;

[0083] Based on the visualization results, the parameters of the quality change standard function and the genetic algorithm can be manually adjusted to optimize the generation process of the storage environment condition scheme.

[0084] According to the final target environmental parameters, the storage environment of biochemical materials is regulated and controlled, and the operating parameters of relevant environmental control equipment (such as ventilation systems, dehumidifiers, air conditioners, heating devices, etc.) are adjusted to ensure that the materials are stored in an environment that meets quality requirements and has the lowest energy consumption.

[0085] Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0086] In the embodiments provided by the present invention, it should be understood that the disclosed systems or methods can be implemented in other ways. For example, the embodiments of the invention described above are merely illustrative. For example, the division of modules is only a logical function division, and other division methods may be used in actual implementation.

[0087] Modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical modules, and may be located in one place or distributed across multiple network modules. Some or all of these modules may be selected to achieve the objectives of this embodiment based on actual needs.

[0088] In addition, the functional modules in various embodiments of the present invention may be integrated into a single processing module, each module may exist physically separately, or two or more modules may be integrated into a single module. The aforementioned integrated modules may be implemented in the form of hardware or hardware plus software functional modules.

[0089] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the basic characteristics of the present invention.

[0090] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A biochemical material storage system based on the Internet of Things, characterized in that: include: Real-time monitoring module: Through sensors, it collects storage environment data and material quality change data of biochemical materials in real time, and obtains energy consumption data, and transmits the storage environment data, material quality change data, and energy consumption data to the data storage module in real time; Data storage module: stores data from the real-time monitoring module; Quality change prediction and energy consumption prediction model generation module: Analyzes the storage environment data, material quality change data, and energy consumption data of the data storage module to build quality change prediction and energy consumption prediction models; Storage environment condition plan generation module: For a certain biochemical material, a quality change standard function is set. Based on historical environmental parameters, quality change prediction and energy consumption prediction models are used to obtain multiple sets of environmental parameters that are lower than the quality change standard function. Then, through data analysis, a set of target environmental parameters is obtained with the goal of minimizing energy consumption. Through sensors, the storage environment data and material quality change data of biochemical materials are collected in real time, and energy consumption data is obtained, and the storage environment data, material quality change data, and energy consumption data are transmitted to the data storage module in real time; Extract storage environment data, material quality change data and energy consumption data from the data storage module, use support vector regression algorithm to build a quality change curve prediction model and an energy consumption prediction model, integrate the quality change curve prediction model and the energy consumption prediction model to form a quality change prediction and energy consumption prediction model, and after receiving the environmental parameter input, the quality change prediction and energy consumption prediction models respectively output the quality change curve prediction result and energy consumption prediction result per unit time; For a certain biochemical material, a quality change standard function is set. Based on historical environmental parameters, multiple groups of environmental parameters that are lower than the quality change standard function are obtained through quality change prediction and energy consumption prediction models. These groups are used as the initial population. The energy consumption of each environmental parameter in the initial population is calculated through the quality change prediction and energy consumption prediction models. X environmental parameters with the lowest energy consumption are selected. These X parameters are cross-mutated and mutated using a genetic algorithm to generate new parameters. The quality change curve of the new parameters is then calculated using the quality change prediction and energy consumption prediction models. The remaining environmental parameters that are lower than the quality change standard function are merged with the initial population to form a new population, and multiple rounds of iterations are performed: In the i-th iteration, i = 1, 2, ..., n, for X environmental parameters, obtain the corresponding X energy consumption values ​​and X quality change curves, find the maximum energy consumption value and record it as E, for the environmental parameter in the iteration group, if its energy consumption is not greater than 1.02E and its quality change curve is lower than any of the X quality change curves, then bring it into the crossover and mutation operation of the X environmental parameters; Continue iterating until the absolute value of the difference in the lowest energy consumption obtained in Y consecutive iterations is less than the set threshold. The goal is to minimize energy consumption, and finally select the environmental parameter with the lowest energy consumption as the target environmental parameter. During the iteration process, the energy consumption values ​​and quality change curve values ​​corresponding to the environmental parameters in the current group are visualized in real time, and the relationship between them is displayed through a two-dimensional scatter plot or Pareto frontier diagram. Based on the visualization results, the parameters of the quality change standard function and the genetic algorithm can be manually adjusted; The storage environment of biochemical materials is regulated based on the final determined target environmental parameters.

2. The biochemical material storage system based on the Internet of Things according to claim 1 is characterized in that: Sensors include temperature sensors, humidity sensors, pressure sensors, gas sensors, material weight sensors, and liquid level sensors.

3. The biochemical material storage system based on the Internet of Things according to claim 1 is characterized in that: Construct quality change prediction and energy consumption prediction models as follows: The storage environment data and material quality change data of the data storage module are analyzed. The quality change curve of a biochemical material under different environmental conditions per unit time is calculated using the support vector regression algorithm. A quality change curve prediction model is constructed. In the quality change curve prediction model, the environmental parameters are used as input, and the quality change curve prediction result per unit time is used as output. The storage environment data and energy consumption data of the data storage module are analyzed. The energy consumption of a certain biochemical material under different environmental conditions per unit time is calculated using the support vector regression algorithm. An energy consumption prediction model is constructed. In the energy consumption prediction model, environmental parameters are used as input, and the energy consumption prediction result per unit time is used as output. The quality change curve prediction model and the energy consumption prediction model are integrated to construct quality change prediction and energy consumption prediction models. When new environmental parameters are input, they are input into the quality change curve prediction model and the energy consumption prediction model respectively. The quality change curve prediction model outputs the prediction result of the quality change curve per unit time, and the energy consumption prediction model outputs the energy consumption prediction result per unit time.

4. The biochemical material storage system based on the Internet of Things according to claim 3 is characterized in that: In the storage environment condition scheme generation module, the target environment parameters are obtained through the following methods: S1. For a certain biochemical material, set a quality change standard function. According to the historical environmental parameters, through the quality change prediction and energy consumption prediction model, obtain multiple groups of environmental parameters lower than the quality change standard function, and set them as the initial group. ; S2. For each environmental parameter that is lower than the quality change standard function, calculate the corresponding energy consumption through the quality change prediction and energy consumption prediction model, and select the X environmental parameters with the lowest energy consumption; S3, X environmental parameters are subjected to crossover and mutation operations through genetic algorithms to obtain multiple new environmental parameters, and the quality change curve of each environmental parameter is calculated to obtain multiple sets of environmental parameters that are lower than the quality change standard function, and the obtained multiple sets of environmental parameters that are lower than the quality change standard function are compared with the Merge to form new groups ; S4. Repeat steps S2 and S3. After multiple rounds of iterations, when the absolute value of the difference in the minimum energy consumption obtained in Y consecutive iterations is less than the set threshold, the environmental parameter with the lowest energy consumption is taken as the target environmental parameter.

5. The biochemical material storage system based on the Internet of Things according to claim 4 is characterized in that: X is a non-negative even number, and 4≤X; Y is a positive integer, and 3≤X.

6. The biochemical material storage system based on the Internet of Things according to claim 5 is characterized in that There are n iterations. For the i-th iteration, i=1,2,...,n, X environmental parameters yield X energy consumption values ​​and X quality change curves. Let the energy consumption value with the largest of the X energy consumption values ​​be E. For the population of the i-th iteration ,group In the process, if there is an environmental parameter whose energy consumption is not greater than 1.02E and whose quality change curve is lower than any of the X quality change curves, it will be brought into the crossover and mutation operation of the X environmental parameters.

7. The biochemical material storage system based on the Internet of Things according to claim 6 is characterized in that: During the iteration process, the energy consumption values ​​and quality change curve values ​​corresponding to the environmental parameters in the current group are visualized in real time to form a two-dimensional scatter plot or Pareto frontier diagram; The quality change standard function and genetic algorithm parameters can be manually adjusted based on the visualization results.

Images