Efficient refrigerating unit control system and method for deep-sea fish

Abstract

The invention discloses an efficient refrigerating unit control system and method for deep-sea fish, and relates to the technical field of equipment control. The method comprises the steps that influence values of all environment data on refrigerating efficiency are calculated, and the environment data influencing the refrigerating efficiency are extracted as key influence factors; extracting two classification nodes, and dividing the key influence factors into a first-level factor, a second-level factor and a third-level factor by using the two classification nodes; all key influence factors and refrigeration parameters of a historical refrigeration unit in work are collected, and a relation mapping model of the key influence factors and the refrigeration parameters is constructed; calculating to obtain a refrigeration parameter adjusting rate corresponding to each grade of key influence factor by utilizing the influence values of different grades of key influence factors; detecting the data value of the key influence factor in real time when the deep-sea fish is stored, and performing judgment and early warning by using a change threshold value; and according to the level of the early warning key influence factor, selecting a corresponding refrigeration parameter adjusting rate to adjust the refrigeration parameter to a target value.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment control, and particularly to an efficient refrigeration unit control system and method for deep-sea fish. Background Technique

[0002] Deep-sea fish live in the deep-sea environment with low temperature and high pressure. After being caught and leaving their original living environment, they are extremely perishable at room temperature. In order to maintain the quality and taste of deep-sea fish and extend their shelf life, it is necessary to quickly reduce their temperature and maintain them at an appropriate low temperature state, which poses high requirements for the performance and control of the refrigeration unit. With the increasing attention to energy and environmental issues, the refrigeration industry is also facing the dual pressures of energy conservation and environmental protection. For deep-sea fish refrigeration units, it is necessary to reduce energy consumption as much as possible and reduce the impact on the environment while ensuring the refrigeration effect, such as using environmentally friendly refrigerants and optimizing the operating parameters of the refrigeration system. With the continuous expansion of the scale of modern fishery production, the requirements for the automation and intelligent control of the refrigeration system are getting higher and higher. Through an efficient control system, functions such as remote monitoring, fault diagnosis, and automatic adjustment of the refrigeration unit can be realized, improving production efficiency, reducing labor costs, and ensuring the stable operation of the refrigeration system.

[0003] However, in the current refrigeration preservation of deep-sea fish, since the transportation of deep-sea fish often needs to span a long latitude and distance, the change of external environmental data is relatively large during the transportation and preservation process. Currently, the refrigeration parameters are relatively fixed during refrigeration preservation and cannot automatically adjust the refrigeration parameters according to environmental changes. When manually adjusting the refrigeration parameters, the adjustment rate is relatively arbitrary, greatly damaging the preservation quality of deep-sea fish. Summary of the Invention

[0004] The purpose of the present invention is to provide an efficient refrigeration unit control system and method for deep-sea fish to solve the problems raised in the prior art.

[0005] To achieve the above purpose, the present invention provides the following technical solutions:

[0006] An efficient refrigeration unit control method for deep-sea fish, the method comprising the following steps:

[0007] S100. Collect all environmental data and refrigeration efficiency during the operation of the historical refrigeration unit, calculate the influence value of all environmental data on the refrigeration efficiency, and extract the environmental data that affects the refrigeration efficiency as the key influencing factors;

[0008] Further, the specific steps of extracting the environmental data that affects the refrigeration efficiency as the key influencing factors are:

[0009] S101. Collect all environmental data and refrigeration efficiency during the operation of the historical refrigeration unit. Take the refrigeration efficiency as the dependent variable and the environmental data as the independent variables to construct a multiple linear regression model of environmental data and refrigeration efficiency. The model formula is:

[0010] y = β0 + β1×x1 + β2×x2 +... + β n ×x n ;

[0011] In the formula, y represents the refrigeration efficiency, x1, x2,... x n represent the first, second,... nth types of environmental data, β0 represents the constant term in the model, and β1, β2,... β n represent the coefficients of the first, second,... nth types of environmental data;

[0012] S102. Collect all environmental data and refrigeration efficiency during the operation of the historical refrigeration unit m times as observation values. Let the observation value of the ith type of environmental data be {x i1 、x i2 、x i3 、... x im}, where x i1 、x i2 、x i3 、... x im represent the values of the ith type of environmental data collected in the 1st, 2nd, 3rd,... mth times. Use all the observation values of the n types of environmental data to construct an independent variable observation matrix. Set the first column of the observation matrix to 1, corresponding to the constant term in the multiple linear regression model. The number of elements in the independent variable observation matrix is m×(n + 1); Use the m collected refrigeration efficiencies to construct a variable observation matrix. The number of elements in the variable observation matrix is m; Calculate the coefficient matrix using the variable observation matrix and the independent variable observation matrix. The formula is:

[0013] β′ = (X T X) -1 X T Y;

[0014] In the formula, β’ represents the coefficient matrix, X represents the independent variable observation matrix, X T represents the transpose matrix of the independent variable observation matrix, and Y represents the variable observation matrix; β’ = [β0, β1, β2,... β n T ;

[0015] S103. Extract β1, β2,... β n in the coefficient matrix as the influence values of the corresponding types of environmental data. Manually set the influence value threshold to β threshold . When |β i | > β threshold ​When it is determined that the corresponding environmental data is the key influencing factor G, after judging the influence values of all environmental data, a set of key influencing factors is obtained.

[0016] By collecting all environmental data and refrigeration efficiency during the operation of the historical refrigeration unit, various factors that may affect the refrigeration efficiency can be comprehensively grasped, avoiding omission of important information. Calculating the influence values of all environmental data on the refrigeration efficiency and extracting the key influencing factors helps to clarify the key objects of attention, provides a basis for subsequent optimization of the refrigeration system, and can targetedly control and adjust these key factors to improve the refrigeration efficiency.

[0017] S200. Analyze the influence values of all key influencing factors, extract two classification nodes, and use the two classification nodes to divide the key influencing factors into primary factors, secondary factors, and tertiary factors;

[0018] Furthermore, the specific steps of using the two classification nodes to divide the key influencing factors into primary factors, secondary factors, and tertiary factors are as follows:

[0019] S201. Sort the absolute values |β i | of all influence values in the set of key influencing factors from small to large, calculate the absolute value differences of adjacent influence values, compare and judge all differences, and extract the maximum difference value and the second largest difference value. Let the absolute values of the two influence values corresponding to the maximum difference value be [β before ,β after , and extract a classification node J1 from β before ; similarly, extract a classification node J2 from the second largest difference value.

[0020] S202. Screen the two extracted classification nodes to obtain the primary node and the secondary node. The formula is:

[0021] J frist =max{J1, J2}

[0022] J second =min{J1, J2};

[0023] In the formula, J first represents the primary node, and J second represents the secondary node; use the two nodes to judge all key influencing factors. When |β i |<J second , it is judged as a tertiary factor; when J second ≤|β i |≤J first , it is judged as a secondary factor; when |β i |>J firstWhen it is, it is judged as a first-level factor; after all the key influencing factors in the set of key influencing factors are judged, three levels of factor sets are obtained.

[0024] Analyze the influence values of the key influencing factors and extract classification nodes, and classify the key influencing factors into different levels, which is conducive to more detailed management and research of the factors. Different levels of factors may require different management strategies and control measures, which is convenient for more targeted optimization.

[0025] S300. Extract the equipment parameters of the refrigeration unit during refrigeration as refrigeration parameters, collect all the key influencing factors and refrigeration parameters of the historical refrigeration unit during operation, and construct a relationship mapping model between the key influencing factors and the refrigeration parameters;

[0026] Furthermore, the specific steps for constructing the relationship mapping model between the key influencing factors and the refrigeration parameters are as follows:

[0027] S301. Extract the equipment parameters of the refrigeration unit during refrigeration as refrigeration parameters, collect all the key influencing factors G and refrigeration parameters P of the historical refrigeration unit during operation; divide the collected key influencing factors and refrigeration parameters into a training set, a validation set and a test set; standardize all the collected data;

[0028] S302. Construct the structure of the mapping model, specifically: set the input layer, the number of input layer nodes is the number of changes in the collected key influencing factors, set the output layer, the number of output layer nodes is the number of collected refrigeration parameters, set the hidden layer, and manually set the number of hidden layer nodes; use ReLU in the hidden layer and a linear activation function in the output layer; use machine learning to train the mapping model to obtain a specific relationship mapping model, and the formula is:

[0029] P j = f(G1, G2,..., G k ; W);

[0030] In the formula, P j represents the jth type of refrigeration parameter, G1, G1,..., G1 represent the 1st, 2nd,..., kth types of key influencing factors, and W represents the model weight parameter; the model weight parameter is obtained by machine training fitting.

[0031] Collect the key influencing factors and refrigeration parameters and construct a relationship mapping model, which can clearly reveal the internal connection between the key influencing factors and the refrigeration parameters. This helps to deeply understand the operation mechanism of the refrigeration system and provides theoretical support for subsequent parameter adjustment and optimization. Through this model, the change trend of the refrigeration parameters can be predicted according to the changes of the key influencing factors, so as to adjust the parameters in advance and realize the optimal control of the refrigeration system, improving the refrigeration effect and efficiency.

[0032] S400. Calculate the corresponding adjustment rate of the refrigeration parameter for each level of the key influencing factor based on the influence values of different levels of key influencing factors.

[0033] Further, the specific steps for calculating the corresponding adjustment rate of the refrigeration parameter for each level of the key influencing factor based on the influence values of different levels of key influencing factors are as follows:

[0034] S401. Collect the records of abnormalities that occurred during the refrigerated storage of deep-sea fish in history, extract the key influencing factor with the largest change in each record, mark it as the abnormal record caused by the corresponding key influencing factor, judge and count the key influencing factors in the records to obtain the total number of abnormal records within each level, and calculate the level weight of the key influencing factor for each level. The formula is:

[0035]

[0036] In the formula, Le v represents the level weight of the key influencing factor of the v-th level, C v represents the total number of abnormal records within the v-th level, C z represents the total number of all collected abnormal records; calculate the level weights of the key influencing factors of the three levels in sequence.

[0037] S402. Calculate the adjustment rate of the refrigeration parameter for the key influencing factor of each level. The formula is:

[0038]

[0039] In the formula, V v represents the adjustment rate of the refrigeration parameter of the key influencing factor of the v-th level, |βp| v represents the average value of the influence values of all key influencing factors within the v-th level, represents the sum of the influence values of all key influencing factors within the v-th level; calculate the adjustment rates of the refrigeration parameters of the key influencing factors of the three levels as V1, V2, and V3 respectively in the same way.

[0040] Calculating the corresponding adjustment rate of the refrigeration parameter for each level of the key influencing factor can accurately adjust the refrigeration parameter according to the influence degree of different levels of factors. It avoids problems such as energy waste and poor refrigeration effect caused by blind parameter adjustment, and realizes the precision and intelligence of refrigeration parameter adjustment. A reasonable adjustment rate helps to maintain the stability of the refrigeration system, prevent the system from being impacted due to too fast or too slow parameter adjustment, and extend the service life of the refrigeration equipment.

[0041] S500. Collect the data values of the key influencing factors when the refrigerated preservation of historical deep-sea fish fails, and calculate the change thresholds for each key influencing factor respectively; during the preservation of deep-sea fish, detect the data values of the key influencing factors in real time and use the change thresholds to make judgment and early warning;

[0042] Further, the specific steps for detecting the data values of the key influencing factors in real time during the preservation of deep-sea fish and using the change thresholds to make judgment and early warning are as follows:

[0043] S501. Collect the data values of the key influencing factors when the refrigerated preservation of historical deep-sea fish fails, calculate the average value and standard deviation of the data values of the key influencing factors when the refrigerated preservation of historical deep-sea fish fails respectively, and calculate the change threshold using the average value and standard deviation. The formula is:

[0044] Th = x_p ± x_b;

[0045] In the formula, Th represents the change threshold of the key influencing factor, x_p represents the average value of the data values of the key influencing factor, and x_b represents the standard deviation of the data values of the key influencing factor; when the influence value of the key influencing factor is negative, the symbol of the threshold formula is selected as +, and when the influence value of the key influencing factor is positive, the symbol of the threshold formula is selected as -;

[0046] S502. During the preservation of deep-sea fish, detect the data values of all key influencing factors in real time as xs, and use the change threshold to make a judgment. When the influence value of the key influencing factor is negative, the judgment rule is: when xs ≥ Th, it is judged that there is an abnormal risk in the refrigerated preservation and early warning is given; when xs < Th, it is judged that the refrigerated preservation is normal;

[0047] When the influence value of the key influencing factor is positive, the judgment rule is opposite.

[0048] S600. When the refrigeration unit gives an early warning, collect the warning key influencing factors and input them into the relationship mapping model to calculate the adjustment target value of the refrigeration parameters; select the corresponding refrigeration parameter adjustment rate according to the level of the warning key influencing factor and adjust the refrigeration parameters to the target value.

[0049] Further, the specific steps for selecting the corresponding refrigeration parameter adjustment rate according to the level of the warning key influencing factor and adjusting the refrigeration parameters to the target value are as follows:

[0050] S601. When the refrigeration unit gives an early warning, collect the warning key influencing factors and input them into the relationship mapping model to calculate the adjustment target value Pm of the refrigeration parameters, search for the real-time warning key influencing factors in the three-level factor set, and obtain the level of the real-time warning key influencing factor;

[0051] Adjust the refrigeration parameters in the refrigeration unit by selecting the corresponding adjustment rate of the refrigeration parameters according to the level of the key influencing factors for real-time warning until the adjustment target value Pm is reached and then stop.

[0052] When the refrigeration unit issues a warning, calculate the adjustment target value of the refrigeration parameters using the relationship mapping model, and select the corresponding adjustment rate according to the level of the warning key influencing factors for adjustment. This can quickly and accurately respond to problems in the refrigeration system, promptly adjust the refrigeration parameters to restore the system to a normal state, and ensure the preservation conditions of deep-sea fish. Selecting the adjustment rate according to the factor level further optimizes the adjustment strategy, making the adjustment process more scientific and reasonable. It not only considers the different impacts of different factors on the refrigeration efficiency but also minimizes the interference to the system while ensuring the adjustment effect.

[0053] An efficient refrigeration unit control system for deep-sea fish, the efficient refrigeration unit control system includes a data acquisition module, a key influencing factor search module, an influencing factor classification module, a mapping model construction module, an adjustment rate calculation module, a warning module, and an adjustment module;

[0054] The data acquisition module is used to collect all environmental data and refrigeration efficiency during the operation of the historical refrigeration unit, and collect the records of abnormalities that occurred during the refrigeration preservation of deep-sea fish in history;

[0055] The key influencing factor search module is used to calculate the influence value of all environmental data on the refrigeration efficiency, and extract the environmental data that affects the refrigeration efficiency as the key influencing factors;

[0056] The influencing factor classification module is used to analyze the influence values of all key influencing factors, extract two classification nodes, and use the two classification nodes to divide the key influencing factors into first-level factors, second-level factors, and third-level factors;

[0057] The mapping model construction module is used to collect all key influencing factors and refrigeration parameters during the operation of the historical refrigeration unit, and construct a relationship mapping model between the key influencing factors and the refrigeration parameters;

[0058] The adjustment rate calculation module is used to calculate the corresponding adjustment rate of the refrigeration parameters for each level of key influencing factors using the influence values of different levels of key influencing factors;

[0059] The warning module is used to calculate the change threshold of each key influencing factor and perform real-time judgment and warning;

[0060] The adjustment module is used to select the adjustment rate to adjust the refrigeration parameters according to the level of the warning key influencing factors after the refrigeration unit issues a warning.

[0061] The adjustment rate calculation module includes a level weight calculation unit and an adjustment rate calculation unit;

[0062] The level weight calculation unit is used to judge and count the key influencing factors in the record to obtain the total number of abnormal records in each level, and calculate the level weight of the key influencing factors in each level.

[0063] The adjustment rate calculation unit is used to calculate the corresponding refrigeration parameter adjustment rate of the key influencing factors in each level by using the influence values and level weights of the key influencing factors at different levels.

[0064] The warning module includes a threshold calculation unit and a judgment and warning unit.

[0065] The threshold calculation unit is used to calculate the average value and standard deviation of the key influencing factor data values when the refrigeration preservation of deep-sea fish fails historically, and calculate the change threshold by using the average value and standard deviation.

[0066] The judgment and warning unit is used to perform real-time judgment and warning on each key influencing factor by using the change threshold.

[0067] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0068] 1. By calculating the influence value of environmental data on the refrigeration efficiency, the present invention accurately locates the key influencing factors, and calculates the corresponding refrigeration parameter adjustment rate according to the influence values of the key influencing factors at different levels, realizing the precise adjustment of refrigeration parameters. This helps to minimize energy consumption and improve energy utilization efficiency to achieve the energy-saving goal while meeting the refrigeration requirements.

[0069] 2. The present invention constructs a relationship mapping model between key influencing factors and refrigeration parameters, which can adjust refrigeration parameters in real time according to the changes of key influencing factors, so that the refrigeration system always maintains the best operating state, thereby improving the refrigeration effect. During the preservation of deep-sea fish, by detecting the data values of key influencing factors in real time and using the change threshold for judgment and warning, problems that may affect the refrigeration preservation effect can be discovered and solved in time, which helps to ensure the quality of deep-sea fish and reduce product losses caused by improper refrigeration. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 It is a module distribution diagram of an efficient refrigeration unit control system for deep-sea fish according to the present invention;

[0071] Figure 2 It is a step schematic diagram of an efficient refrigeration unit control method for deep-sea fish according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0072] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0073] Embodiment: As Figure 1 - Figure 2 shown, the present invention provides a technical solution,

[0074] An efficient refrigeration unit control method for deep-sea fish, the method comprising the following steps:

[0075] S100. Collect all environmental data and refrigeration efficiency during the operation of the historical refrigeration unit, calculate the influence value of all environmental data on the refrigeration efficiency, and extract the environmental data that affects the refrigeration efficiency as the key influencing factors;

[0076] The specific steps of extracting the environmental data that affects the refrigeration efficiency as the key influencing factors are:

[0077] S101. Collect all environmental data and refrigeration efficiency during the operation of the historical refrigeration unit, take the refrigeration efficiency as the dependent variable and the environmental data as the independent variables, and construct a multiple linear regression model of the environmental data and the refrigeration efficiency. The model formula is:

[0078] y = β0 + β1×x1 + β2×x2 +... + β n ×x n ;

[0079] In the formula, y represents the refrigeration efficiency, x1, x2,... x n represent the 1st, 2nd,... nth kinds of environmental data, β0 represents the constant term in the model, and β1, β2,... β n represent the coefficients of the 1st, 2nd,... nth kinds of environmental data;

[0080] S102. Collect the all environmental data and refrigeration efficiency during the operation of the historical refrigeration unit m times as the observed values. Let the observed value of the i-th kind of environmental data be {x i1 、x i2 、x i3 、...x im}, where x i1 、x i2 、x i3 、...x imDenote the i-th environmental data value for the 1st, 2nd, 3rd, ..., m-th collection. Use all the observed values of n environmental data to construct an independent variable observation matrix. Set the first column of the observation matrix to 1, corresponding to the constant term in the multiple linear regression model. The number of elements in the independent variable observation matrix is m×(n + 1); use the m collected refrigeration efficiencies to construct a variable observation matrix, and the number of elements in the variable observation matrix is m; calculate the coefficient matrix using the variable observation matrix and the independent variable observation matrix, and the formula is:

[0081] β′=(X T X) -1 X T Y;

[0082] In the formula, β’ represents the coefficient matrix, X represents the independent variable observation matrix, X T represents the transpose matrix of the independent variable observation matrix, and Y represents the variable observation matrix; β’=[β0, β1, β2, ..., β n T ;

[0083] S103. Extract β1, β2, ..., β n from the coefficient matrix as the influence values of the corresponding types of environmental data. Manually set the influence value threshold to β threshold . When |β i | > β threshold , determine that the corresponding environmental data is a key influencing factor G. After judging the influence values of all environmental data, obtain the key influencing factor set.

[0084] By collecting all environmental data and refrigeration efficiency during the operation of the historical refrigeration unit, various factors that may affect the refrigeration efficiency can be comprehensively grasped, avoiding missing important information. Calculating the influence values of all environmental data on the refrigeration efficiency and extracting the key influencing factors helps to clarify the key objects of attention, provides a basis for subsequent optimization of the refrigeration system, and can targetedly control and adjust these key factors to improve the refrigeration efficiency.

[0085] S200. Analyze the influence values of all key influencing factors, extract two classification nodes, and use the two classification nodes to divide the key influencing factors into first-level factors, second-level factors, and third-level factors;

[0086] The specific steps to divide the key influencing factors into first-level factors, second-level factors, and third-level factors using the two classification nodes are as follows:

[0087] S201. For all the absolute values of the influence values |β i ​Sort from small to large, calculate the absolute value difference of adjacent influence values, compare and judge all differences, and then extract the maximum difference and the second largest difference. Let the absolute values of the two influence values corresponding to the maximum difference be [β before , β after . Select a classification node J1 for β before , and select a classification node J2 from the second largest difference in the same way;

[0088] S202. Screen the two extracted classification nodes to obtain the first-level node and the second-level node. The formula is:

[0089] J frist = max{J1, J2}

[0090] J second = min{J1, J2};

[0091] In the formula, J first represents the first-level node, and J second represents the second-level node; use the two types of nodes to judge all key influence factors. When |β i | < J second , it is judged as a third-level factor; when J second ≤ |β i | ≤ J first , it is judged as a second-level factor; when |β i | > J first , it is judged as a first-level factor; after judging all the key influence factors in the key influence factor set, three levels of factor sets are obtained.

[0092] Analyze the influence values of the key influence factors and extract classification nodes, which divides the key influence factors into different levels, facilitating more detailed management and research of the factors. Factors at different levels may require different management strategies and control measures, making it easier to optimize them more targeted.

[0093] S300. Extract the equipment parameters of the refrigeration unit during refrigeration as refrigeration parameters, collect all the key influence factors and refrigeration parameters of the historical refrigeration unit during operation, and construct a relationship mapping model between the key influence factors and the refrigeration parameters;

[0094] The specific steps for constructing the relationship mapping model between the key influence factors and the refrigeration parameters are as follows:

[0095] S301. Extract the equipment parameters of the refrigeration unit during refrigeration as refrigeration parameters, collect all the key influence factors G and refrigeration parameters P of the historical refrigeration unit during operation; divide the collected key influence factors and refrigeration parameters into training set, validation set and test set; standardize all the collected data;

[0096] S302. Construct the structure of the mapping model, specifically: set the input layer, the number of nodes in the input layer is the number of collected key influencing factor change values, set the output layer, the number of nodes in the output layer is the number of collected refrigeration parameters, set the hidden layer, and manually set the number of nodes in the hidden layer; use ReLU in the hidden layer and a linear activation function in the output layer; use machine learning to train the mapping model to obtain a specific relationship mapping model, and the formula is:

[0097] P j =f(G1, G2,..., G k ; W);

[0098] In the formula, P j represents the jth refrigeration parameter, G1, G1,..., G1 represent the 1st, 2nd,..., kth key influencing factors, and W represents the model weight parameter; the model weight parameter is obtained by machine training and fitting.

[0099] Collecting key influencing factors and refrigeration parameters and constructing a relationship mapping model can clearly reveal the internal connection between key influencing factors and refrigeration parameters. This helps to deeply understand the operation mechanism of the refrigeration system and provides theoretical support for subsequent parameter adjustment and optimization. Through this model, the change trend of refrigeration parameters can be predicted based on the change of key influencing factors, so as to adjust the parameters in advance and realize the optimal control of the refrigeration system, improving the refrigeration effect and efficiency.

[0100] S400. Calculate the corresponding refrigeration parameter adjustment rate for each level of key influencing factors using the influence values of different levels of key influencing factors;

[0101] The specific steps for calculating the corresponding refrigeration parameter adjustment rate for each level of key influencing factors using the influence values of different levels of key influencing factors are as follows:

[0102] S401. Collect the records of abnormalities that occurred during the refrigerated storage of deep-sea fish in history, extract the key influencing factor with the largest change in each record, mark it as the abnormal record caused by the corresponding key influencing factor, judge and count the key influencing factors in the records to obtain the total number of abnormal records within each level, and calculate the level weight of the key influencing factors at each level. The formula is:

[0103]

[0104] In the formula, Le v represents the level weight of the key influencing factor of the vth level, C v represents the total number of abnormal records within the vth level, and C z represents the total number of all collected abnormal records; calculate the level weights of the key influencing factors at three levels in sequence;

[0105] S402. Calculate the adjustment rate of the refrigeration parameters for each level of key influencing factors. The formula is:

[0106]

[0107] In the formula, V v represents the adjustment rate of the refrigeration parameters for the key influencing factor of the v-th level, and |βp| v represents the average value of the influence values of all key influencing factors within the v-th level. represents the sum of the influence values of all key influencing factors within the v-th level. The adjustment rates of the refrigeration parameters for the key influencing factors of the three levels obtained by the same calculation are V1, V2, and V3 respectively.

[0108] Calculating the adjustment rate of the refrigeration parameters corresponding to each level of key influencing factors can accurately adjust the refrigeration parameters according to the influence degree of factors at different levels. It avoids problems such as energy waste and poor refrigeration effect caused by blind parameter adjustment, and realizes the precision and intelligence of refrigeration parameter adjustment. A reasonable adjustment rate helps to maintain the stability of the refrigeration system, prevent the impact on the system caused by too fast or too slow parameter adjustment, and extend the service life of the refrigeration equipment.

[0109] S500. Collect the data values of the key influencing factors when the historical deep-sea fish refrigeration preservation fails, and calculate the change thresholds of each key influencing factor respectively; during the preservation of deep-sea fish, detect the data values of the key influencing factors in real time and use the change thresholds for judgment and early warning;

[0110] The specific steps for detecting the data values of the key influencing factors in real time during the preservation of deep-sea fish and using the change thresholds for judgment and early warning are as follows:

[0111] S501. Collect the data values of the key influencing factors when the historical deep-sea fish refrigeration preservation fails, calculate the average value and standard deviation of the data values of the key influencing factors when the historical deep-sea fish refrigeration preservation fails respectively, and calculate the change threshold using the average value and standard deviation. The formula is:

[0112] Th = x_p ± x_b;

[0113] In the formula, Th represents the change threshold of the key influencing factor, x_p represents the average value of the data values of the key influencing factor, and x_b represents the standard deviation of the data values of the key influencing factor; when the influence value of the key influencing factor is negative, the + sign is selected for the threshold formula symbol, and when the influence value of the key influencing factor is positive, the - sign is selected for the threshold formula symbol;

[0114] S502. When the data values of all key influencing factors are detected in real time during the preservation of deep-sea fish as xs, use the change threshold for judgment. When the influence value of the key influencing factor is negative, the judgment rule is: when xs≥Th, it is judged that there is an abnormal risk in the refrigeration preservation and a warning is issued; when xs<Th, it is judged that the refrigeration preservation is normal;

[0115] When the influence value of the key influencing factor is positive, the judgment rule is opposite.

[0116] S600. When the refrigeration unit issues a warning, collect the warning key influencing factors and input them into the relationship mapping model to calculate the adjustment target value of the refrigeration parameters; select the corresponding refrigeration parameter adjustment rate according to the level of the warning key influencing factor and adjust the refrigeration parameters to the target value.

[0117] The specific steps of selecting the corresponding refrigeration parameter adjustment rate according to the level of the warning key influencing factor and adjusting the refrigeration parameters to the target value are as follows:

[0118] S601. When the refrigeration unit issues a warning, collect the warning key influencing factors and input them into the relationship mapping model to calculate the adjustment target value Pm of the refrigeration parameters. Search for the real-time warning key influencing factors in the three-level factor set to obtain the level of the real-time warning key influencing factors;

[0119] Use the level of the real-time warning key influencing factor to select the corresponding refrigeration parameter adjustment rate to adjust the refrigeration parameters in the refrigeration unit until the adjustment target value Pm is reached and then stop.

[0120] When the refrigeration unit issues a warning, use the relationship mapping model to calculate the adjustment target value of the refrigeration parameters, and select the corresponding adjustment rate according to the level of the warning key influencing factor for adjustment, which can quickly and accurately respond to the problems in the refrigeration system. Adjust the refrigeration parameters in a timely manner to restore the system to the normal state and ensure the preservation conditions of deep-sea fish. Select the adjustment rate according to the factor level, further optimizing the adjustment strategy and making the adjustment process more scientific and reasonable. It not only considers the different impacts of different factors on the refrigeration efficiency but also can minimize the interference to the system while ensuring the adjustment effect.

[0121] An efficient refrigeration unit control system for deep-sea fish, the efficient refrigeration unit control system includes a data acquisition module, a key influencing factor search module, an influencing factor grading module, a mapping model construction module, an adjustment rate calculation module, a warning module, and an adjustment module;

[0122] The data acquisition module is used to collect all environmental data and refrigeration efficiency during the operation of the historical refrigeration unit, and collect the records of abnormalities that occurred during the refrigeration preservation of deep-sea fish in history;

[0123] The key influencing factor searching module is used to calculate the influence values of all environmental data on the refrigeration efficiency, and extract the environmental data that affects the refrigeration efficiency as the key influencing factors;

[0124] The influencing factor grading module is used to analyze the influence values of all key influencing factors, extract two classification nodes, and classify the key influencing factors into primary factors, secondary factors and tertiary factors by using the two classification nodes;

[0125] The mapping model construction module is used to collect all key influencing factors and refrigeration parameters during the operation of the historical refrigeration unit, and construct a relationship mapping model between the key influencing factors and the refrigeration parameters;

[0126] The adjustment rate calculation module is used to calculate the corresponding refrigeration parameter adjustment rate of each level of key influencing factors by using the influence values of different levels of key influencing factors;

[0127] The warning module is used to calculate the change threshold of each key influencing factor and make a real-time judgment and warning;

[0128] The adjustment module is used to select the adjustment rate according to the level of the warning key influencing factor to adjust the refrigeration parameters when the refrigeration unit gives a warning.

[0129] The adjustment rate calculation module includes a level weight calculation unit and an adjustment rate calculation unit;

[0130] The level weight calculation unit is used to judge and count the key influencing factors in the record to obtain the total number of abnormal records in each level, and calculate the level weight of the key influencing factors in each level;

[0131] The adjustment rate calculation unit is used to calculate the corresponding refrigeration parameter adjustment rate of each level of key influencing factors by using the influence values and level weights of different levels of key influencing factors.

[0132] The warning module includes a threshold calculation unit and a judgment warning unit;

[0133] The threshold calculation unit is used to calculate the average value and standard deviation of the key influencing factor data values when the historical deep-sea fish refrigeration preservation fails respectively, and calculate the change threshold by using the average value and standard deviation;

[0134] The judgment warning unit is used to make a real-time judgment and warning on each key influencing factor by using the change threshold.

[0135] Example: During the operation of a certain deep-sea fish refrigeration unit, it is necessary to dynamically adjust the refrigeration parameters according to environmental factors to maintain the preservation effect. Through historical data analysis, the following key influencing factors and their influence coefficients on the refrigeration efficiency have been determined as follows:

[0136] Ambient temperature: 0.9, seawater salinity: 0.8, compressor load: 0.6, humidity: 0.4, air pressure: 0.3; sorted from small to large as 0.1, 0.3, 0.5, 0.7, 0.9; the first-level node is calculated to be 0.6 and the second-level node is 0.4; it is determined that the first-level factors are: ambient temperature, seawater salinity; the second-level factors are: compressor load, humidity; the third-level factor is: air pressure;

[0137] The comprehensive influence values in the three levels are calculated to be 1.7, 1, and 0.3 respectively; let the level weights of the three levels be 0.5, 0.4, and 0.1 respectively; the adjustment rates of the refrigeration parameters for the three levels are calculated to be 0.25, 0.2, and 0.03.

[0138] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present invention. Any reference signs in the claims should not be regarded as limiting the claimed rights.

Claims

1. An efficient refrigeration unit control method for deep-sea fish, characterized in that: The method includes the following steps: S100. Collect all environmental data and refrigeration efficiency during the operation of the historical refrigeration unit, calculate the influence value of all environmental data on the refrigeration efficiency, and extract the environmental data that affects the refrigeration efficiency as the key influencing factors; S200. Analyze the influence values of all key influencing factors, extract two classification nodes, and use the two classification nodes to divide the key influencing factors into primary factors, secondary factors, and tertiary factors; S300. Extract the equipment parameters of the refrigeration unit during refrigeration as the refrigeration parameters, collect all key influencing factors and refrigeration parameters during the operation of the historical refrigeration unit, and construct a relationship mapping model between the key influencing factors and the refrigeration parameters; S400. Calculate the corresponding refrigeration parameter adjustment rate for each level of key influencing factors by using the influence values of different levels of key influencing factors; S500. Collect the data values of the key influencing factors when the historical deep-sea fish refrigeration preservation fails, and calculate the change threshold for each key influencing factor respectively; during the preservation of deep-sea fish, detect the data values of the key influencing factors in real time and use the change threshold for judgment and early warning; S600. When the refrigeration unit issues an early warning, collect the warning key influencing factors and input them into the relationship mapping model to calculate the adjustment target value of the refrigeration parameters; select the corresponding refrigeration parameter adjustment rate according to the level of the warning key influencing factors and adjust the refrigeration parameters to the target value.

2. The control method of an efficient refrigeration unit for deep-sea fish according to claim 1, wherein: The specific steps for extracting the environmental data that affects the refrigeration efficiency as the key influencing factors in S100 are as follows: S101. Collect all environmental data and refrigeration efficiency during the operation of the historical refrigeration unit, take the refrigeration efficiency as the dependent variable and the environmental data as the independent variables, and construct a multiple linear regression model of the environmental data and the refrigeration efficiency. The model formula is: y = β0 + β1×x1 + β2×x2 + … + β n ×x n ; In the formula, y represents the refrigeration efficiency, x1, x2,... x n represent the first, second,... nth environmental data, β0 represents the constant term in the model, β1, β2,... β n represent the coefficients of the first, second,... nth environmental data; S102. Collect all environmental data and refrigeration efficiency of the historical refrigeration unit during operation for m times as observation values. Let the observation value of the i-th environmental data be {x i1 、x i2 、x i3 、...x im}, where x i1 、x i2 、x i3 、...x im represent the values of the i-th environmental data collected in the 1st, 2nd, 3rd,... m-th times. Use all the observation values of n environmental data to construct an independent variable observation matrix. Set the first column of the observation matrix to 1, corresponding to the constant term in the multiple linear regression model. The number of elements in the independent variable observation matrix is m×(n + 1); Use the refrigeration efficiency collected for m times to construct a variable observation matrix, and the number of elements in the variable observation matrix is m; Calculate the coefficient matrix by using the variable observation matrix and the independent variable observation matrix. The formula is: β′ = (X T X) -1 X T Y; In the formula, β’ represents the coefficient matrix, X represents the independent variable observation matrix, and X T represents the transpose matrix of the independent variable observation matrix, and Y represents the variable observation matrix; β’＝[β0、β1、β2、...、β n ] T ； S103. Extract β1, β2, …, β in the coefficient matrix n as the influence values of the corresponding types of environmental data, and manually set the influence value threshold to β threshold , when |β i | > β threshold , it is determined that the corresponding environmental data is a key influencing factor G. After judging the influence values of all environmental data, a set of key influencing factors is obtained.

3. The control method of an efficient refrigeration unit for deep-sea fish according to claim 2, characterized in that: The specific steps for using the two classification nodes to divide the key influencing factors into primary factors, secondary factors, and tertiary factors in S200 are as follows: S201. Sort all the absolute values of the influence values in the set of key influencing factors |β i | from smallest to largest, calculate the absolute value differences between adjacent influence values, compare and judge all the differences, and then extract the maximum difference and the second largest difference. Let the absolute values of the two influence values corresponding to the maximum difference be [β before , β after . Extract a classification node J1 from β before , and extract a classification node J2 from the second largest difference in the same way; S202. Screen the two extracted classification nodes to obtain the primary node and the secondary node. The formula is: J first = max{J1, J2} J second = min{J1, J2}; In the formula, J first represents the primary node, and J second represents the secondary node; all key influencing factors are judged using the two types of nodes. In the set of key influencing factors, when |β i | < J second , it is judged as a third-level factor; when J second ≤ |β i | ≤ J first , it is judged as a second-level factor; when |β i | > J first , it is judged as a first-level factor; after judging all the key influencing factors in the set of key influencing factors, three sets of factors at different levels are obtained.

4. The control method of an efficient refrigeration unit for deep-sea fish according to claim 3, characterized in that: The specific steps for constructing the relationship mapping model between the key influencing factors and the refrigeration parameters in S300 are as follows: S301. Extract the equipment parameters of the refrigeration unit during refrigeration as the refrigeration parameters, collect all key influencing factors G and refrigeration parameters P during the operation of the historical refrigeration unit; divide the collected key influencing factors and refrigeration parameters into a training set, a validation set, and a test set; standardize all the collected data; S302. Construct the structure of the mapping model. Specifically: set the input layer, the number of input layer nodes is the number of change values of the collected key influencing factors, set the output layer, the number of output layer nodes is the number of collected refrigeration parameters, set the hidden layer, and manually set the number of hidden layer nodes; use ReLU in the hidden layer and use a linear activation function in the output layer; use machine learning to train the mapping model to obtain the specific relationship mapping model. The formula is: P j = f(G1, G2,..., G k ; W); In the formula, P j represents the jth refrigeration parameter, G1, G1, ..., G1 represent the 1st, 2nd, ..., kth key influencing factors, and W represents the model weight parameter; the model weight parameter is obtained by machine training and fitting.

5. The control method of an efficient refrigeration unit for deep-sea fish according to claim 4, characterized in that: The specific steps for calculating the corresponding refrigeration parameter adjustment rate for each level of key influencing factors by using the influence values of different levels of key influencing factors in S400 are as follows: S401. Collect the records of abnormalities that occurred during the refrigerated storage of deep-sea fish in history, extract the key influencing factors with the largest changes in each record, mark them as abnormal records caused by the corresponding key influencing factors, judge and count the key influencing factors in the records to obtain the total number of abnormal records in each level, and calculate the level weights of the key influencing factors in each level. The formula is: In the formula, Le v represents the level weight of the key influencing factor of the v-th level, and C v represents the total number of abnormal records within the v-th level, and C z represents the total number of all abnormal records collected; the level weights of the key influencing factors of the three levels are calculated in sequence; S402. Calculate the adjustment rate of the refrigeration parameters for each level of key influencing factors. The formula is: In the formula, V v represents the refrigeration parameter adjustment rate of the key influencing factor at the v-th level, and |βp| v represents the average value of the influence values of all key influencing factors within the v-th level, represents the sum of the influence values of all key influencing factors within the v-th level; the adjustment rates of the refrigeration parameters of the key influencing factors at the three levels are obtained by the same calculation as V1, V2, and V3 respectively.

6. The control method of an efficient refrigeration unit for deep-sea fish according to claim 5, characterized in that: The specific steps for the S500 to detect the data values of the key influencing factors in real time during the preservation of deep-sea fish and use the change threshold for judgment and warning are as follows: S501. Collect the data values of the key influencing factors when the refrigerated storage of historical deep-sea fish fails. Calculate the average value and standard deviation of the data values of the key influencing factors when the refrigerated storage of historical deep-sea fish fails, and use the average value and standard deviation to calculate the change threshold. The formula is: Th = x_p ± x_b; In the formula, Th represents the change threshold of the key influencing factor, x_p represents the average value of the data values of the key influencing factor, and x_b represents the standard deviation of the data values of the key influencing factor; when the influence value of the key influencing factor is negative, the + sign is selected for the threshold formula symbol, and when the influence value of the key influencing factor is positive, the - sign is selected for the threshold formula symbol; S502. During the preservation of deep-sea fish, the data values of all key influencing factors are detected in real time as xs, and the change threshold is used for judgment. When the influence value of the key influencing factor is negative, the judgment rule is: when xs ≥ Th, it is judged that there is an abnormal risk in the refrigerated storage and a warning is issued; when xs < Th, it is judged that the refrigerated storage is normal; When the influence value of the key influencing factor is positive, the judgment rule is the opposite.

7. An efficient refrigeration unit control method for deep-sea fish according to claim 6, characterized in that: The specific steps for the S600 to select the corresponding refrigeration parameter adjustment rate according to the level of the warning key influencing factor and adjust the refrigeration parameters to the target value are as follows: S601. When the refrigeration unit issues a warning, collect the warning key influencing factor and input it into the relationship mapping model to calculate the adjustment target value Pm of the refrigeration parameter. Search for the real-time warning key influencing factor in the set of three-level factors to obtain the level of the real-time warning key influencing factor; Use the level of the real-time warning key influencing factor to select the corresponding refrigeration parameter adjustment rate to adjust the refrigeration parameters in the refrigeration unit until the adjustment target value Pm is reached and then stop.

8. An efficient refrigeration unit control system for deep-sea fish, characterized in that: The high-efficiency refrigeration unit control system includes a data collection module, a key influencing factor search module, an influencing factor classification module, a mapping model construction module, an adjustment rate calculation module, a warning module, and an adjustment module; The data collection module is used to collect all environmental data and refrigeration efficiency during the operation of the historical refrigeration unit, and collect the records of abnormalities that occurred during the refrigerated storage of deep-sea fish in history; The key influencing factor search module is used to calculate the influence values of all environmental data on the refrigeration efficiency, and extract the environmental data that affects the refrigeration efficiency as the key influencing factors; The influencing factor classification module is used to analyze the influence values of all key influencing factors, extract two classification nodes, and use the two classification nodes to divide the key influencing factors into first-level factors, second-level factors, and third-level factors; The mapping model construction module is used to collect all key influencing factors and refrigeration parameters during the operation of the historical refrigeration unit, and construct a relationship mapping model between the key influencing factors and the refrigeration parameters; The adjustment rate calculation module is used to calculate the corresponding refrigeration parameter adjustment rate of each level of key influencing factors by using the influence values of different levels of key influencing factors; The warning module is used to calculate the change threshold of each key influencing factor and make real-time judgment and warning; The adjustment module is used to adjust the refrigeration parameters according to the adjustment rate selected according to the level of the warning key influencing factor when the refrigeration unit gives a warning.

9. The control system of an efficient refrigeration unit for deep-sea fish according to claim 8, characterized in that: The adjustment rate calculation module includes a level weight calculation unit and an adjustment rate calculation unit; The level weight calculation unit is used to judge and count the key influencing factors in the record to obtain the total number of abnormal records in each level, and calculate the level weight of the key influencing factors of each level; The adjustment rate calculation unit is used to calculate the corresponding refrigeration parameter adjustment rate of each level of key influencing factors by using the influence values and level weights of different levels of key influencing factors.

10. The control system of an efficient refrigeration unit for deep-sea fish according to claim 8, characterized in that: The warning module includes a threshold calculation unit and a judgment and warning unit; The threshold calculation unit is used to calculate the average value and standard deviation of the key influencing factor data values respectively when the historical deep-sea fish refrigeration preservation fails, and calculate the change threshold by using the average value and standard deviation; The judgment and warning unit is used to make real-time judgment and warning of each key influencing factor by using the change threshold.

Images