Intelligent control method for plasma instant freezer with combination of multiple refrigeration units

Abstract

The invention relates to the technical field of plasma instant freezers, in particular to an intelligent control method of a multi-refrigeration-unit combined plasma instant freezer, which comprises the following steps of: sequentially starting a plurality of refrigeration units in a delayed manner after electrifying, and operating the refrigeration units at full power after starting; the number of the placed blood bags is collected according to a weight sensor; acquiring load capacity and load temperature data in the instant freezer in real time; generating a rotating speed adjusting instruction by adopting a PID control algorithm based on fuzzy logic; when the load is increased or the temperature rises, the control unit increases the rotating speed of the refrigeration unit; when the load is reduced or the temperature is close to a set threshold value, the control unit reduces the rotating speed of the refrigeration unit; dynamically distributing the work cycle and the start-stop sequence of each refrigeration unit; and through a closed-loop feedback mechanism, the rotating speed adjusting parameters are dynamically corrected according to the actual refrigeration effect. According to the method, high-precision temperature tracking of a fuzzy PID control algorithm is combined, and it is ensured that the biological activity of plasma is not damaged in the quick freezing process.

Description

Technical Field

[0001] The present invention relates to the technical field of plasma quick-freezing machines, in particular to an intelligent control method for a plasma quick-freezing machine with a multi-refrigeration unit combination. Background Art

[0002] Plasma quick freezers are widely used in medical laboratories, blood stations, biological product production and other fields. Their main function is to quickly freeze plasma and extend the shelf life of blood products. They are key equipment in the blood cold chain. Through rapid and uniform ultra-low temperature freezing technology, they ensure the biological activity and clinical value of plasma. They are important supporting tools in modern transfusion medicine and biopharmaceuticals. Their design must take into account efficiency, safety and compliance to ensure that the quality of the entire process from blood collection to use is controllable.

[0003] The invention with publication number CN102853627B discloses a plasma quick-freezing machine, in which the refrigeration plate is divided into a fixed refrigeration plate and a movable refrigeration plate. The fixed refrigeration plate is connected to the inner wall of the box cover, and the movable refrigeration plate is arranged below the fixed refrigeration plate via a lifting mechanism. The refrigerator is connected to the refrigeration plate via a cooling pipe. A plasma tray is provided on the movable refrigeration plate. Fixed guide rails with a raised middle portion are provided on both sides of the movable refrigeration plate. Mobile guide rails of the same shape as the fixed guide rails are provided on the plasma tray corresponding to the fixed guide rails. Guide support wheels are provided on the platforms at the front end of the fixed guide rails and the rear end of the movable guide rails. A displacement sensor is provided above the box cover via a bracket. A position sensor is used to detect the deformation of the box sheet metal to replace the pressure sensor. The displacement sensor and the operation button are respectively connected to the controller via data cables, and the control circuit board is connected to the motor and the refrigerator.

[0004] While this invention boasts advantages such as easy assembly, reduced production costs, simple operation, stable movement, fast freezing speed, large freezing capacity, high degree of automation, and excellent plasma freezing results, it suffers from the following issues: First, the refrigeration power is fixed and cannot be dynamically adjusted according to load changes, resulting in energy waste; second, the startup current surge is large, affecting power grid stability and equipment life; and finally, the temperature control accuracy is insufficient, affecting the quality of plasma freezing. Summary of the Invention

[0005] The object of the present invention is to provide an intelligent control method for a plasma quick-freezing machine with a multi-refrigeration unit combination to solve the problems raised in the above background technology.

[0006] In order to solve the above technical problems, the present invention aims to provide an intelligent control method for a plasma quick-freezing machine with multiple refrigeration units, comprising the following steps: S1: After power-on, several refrigeration units are started sequentially and delayed, and then the refrigeration units are operated at full power; S2. Optimize the refrigeration unit that needs to be turned on based on the number of blood bags placed in the blood bag according to the weight sensor; S3, collecting the load quantity and load temperature data in the quick freezer in real time through the weight sensor and temperature sensor; S4, inputting the load amount and load temperature data into the control unit, and using the PID control algorithm based on fuzzy logic to generate a speed adjustment instruction; S5. When the load increases or the temperature rises, the control unit increases the speed of the refrigeration unit; when the load decreases or the temperature approaches a set threshold, the control unit decreases the speed of the refrigeration unit; S6. Based on the speed adjustment result, dynamically allocate the working cycle and start-stop sequence of each refrigeration unit; S7, through the closed-loop feedback mechanism, dynamically correct the speed adjustment parameters according to the actual cooling effect; S8. Use historical data to establish a refrigeration efficiency feature library.

[0007] As a further improvement of the present technical solution, the delay start time in S1 is 3-10 seconds, and the specific duration is determined by the grid voltage fluctuation threshold.

[0008] As a further improvement of the present technical solution, in S4, the load quantity and load temperature data are input into the control unit, and a PID control algorithm based on fuzzy logic is used to generate a speed adjustment instruction. The specific steps are as follows: S4.1, load preset parameters, initialize PID parameters and fuzzy rule weights, and user inputs set temperature ; S4.2, data collection, temperature error and temperature error rate of change As an input variable, the temperature error To set the temperature With actual temperature The difference, set temperature Set by the user, actual temperature The data measured by the temperature sensor, the temperature error change rate is the temperature error The differential value of , the corresponding formula is: ; ; Among them, the temperature error The domain of , temperature error change rate The domain of ; S4.3, fuzzy processing, the temperature error and temperature error rate of change Map to fuzzy sets and calculate its membership in each fuzzy set; S4.4. Establish rules based on expert opinions; S4.5, matching fuzzy rules, according to the fuzzy set combination of input variables, searching the corresponding output fuzzy set from the fuzzy rule base; S4.6, defuzzification calculation, using the centroid method to convert the fuzzy output into an exact value, the corresponding formula is: ; in, To adjust the speed gradient accurately, For the The membership of the fuzzy set is output by the rules. For the The output value corresponding to the rule; S4.7, output refrigeration unit speed adjustment value, the corresponding formula is: ; in, is the speed adjustment value of the refrigeration unit; S4.8, integral separation anti-saturation, add integral separation strategy, if , freeze the integral term .

[0009] As a further improvement of this technical solution, in S6, the speed adjustment amount of the refrigeration unit is adjusted according to the speed of the refrigeration unit. Allocate work cycles to each unit , the corresponding formula is: ; in, is the maximum permissible speed of the refrigeration unit; The cooling unit adopts polling scheduling.

[0010] As a further improvement of the present technical solution, in S7, the speed adjustment parameter is dynamically corrected according to the actual cooling effect through a closed-loop feedback mechanism. The specific steps include: S7.1. Update the fuzzy rule weights according to the gradient descent method; S7.2. Proportional coefficient based on improved whale algorithm , integral coefficient and differential coefficients Perform optimization.

[0011] As a further improvement of the present technical solution, in S7.1, the fuzzy rule weights are updated according to the gradient descent method, and the corresponding formula is: ; in, is the updated fuzzy rule weight, is the fuzzy rule weight before updating, is the learning rate, is the theoretical cooling rate, is the actual cooling rate, is the partial derivative of the speed regulation of the refrigeration unit with respect to the fuzzy rule weight, Through chain rule calculation, the rule weights can also be restricted to prevent the rules from being over-strengthened or weakened.

[0012] As a further improvement of the present technical solution, in S7.1, the fuzzy rule weights are updated according to the gradient descent method, and the corresponding multi-objective optimization function is: ; in, is a multi-objective optimization function, 、 and are weight coefficients, is the absolute value of temperature deviation, is the refrigeration unit power, It is the system stabilization time.

[0013] As a further improvement of this technical solution, in S7.2, based on the improved whale algorithm, the proportional coefficient , integral coefficient and differential coefficients The specific steps for optimization are: A1. Initialize parameters, set the population size, spatial dimension to three, maximum number of iterations, and initial whale population; A2. Calculate the fitness value. Calculate the fitness value according to the fitness function. The fitness function is: ; in, is the fitness function, is the time-weighted integrated absolute error, is the running time; A3. Boundary condition processing: process each individual in the population and calculate the coefficient vector and , and generate uniformly distributed decision random numbers , the corresponding formula is: ; ; ; in, is a random vector on [0,1], is the control vector, is the initial value of the control vector, is the current iteration number, is the maximum number of iterations; A4. Randomly search for prey and introduce Lecy flight: and When , the corresponding formula is: ; in, is the position vector of a randomly selected individual in the current population, For the current whale The position in the iteration, For the current whale The position in the iteration, is the random step size to obey; A5. Shrink and surround the prey, introduce suboptimal and third-optimal solutions: and When , the corresponding formula is: ; in, 、 and Respectively The optimal, second-optimal, and third-optimal position vectors of the objective function in the population in the iteration; A6. Spiral update position: When hour, is the distance between the current whale and the current optimal whale position, and the corresponding formula is: ; ; in, are the coefficients used to define the shape of the logarithmic spiral, is a uniformly distributed random number between [-1,1]. is the coefficient The initial value of is the decay rate parameter; A7. Determine the termination condition. Stop the iteration when the maximum number of iterations is reached. Otherwise, repeat steps A3 to A6.

[0014] As a further improvement of the present technical solution, in said S7, the triggering conditions for dynamic correction include periodic correction and event-driven correction. The periodic correction is to perform parameter fine-tuning every 30 seconds. The event-driven correction includes the change in the number of blood bags exceeding 2 bags, the temperature difference of the box exceeding 0.5°C, and the compressor current fluctuation in the refrigeration unit exceeding 15%. In said S7, a safety constraint mechanism is also introduced. The safety constraint mechanism includes speed change rate limit and temperature protection threshold. The speed change rate limit is ,in, The temperature protection threshold includes the maximum cooling rate limit and overcooling protection. The maximum cooling rate limit is less than or equal to 3℃ / min to prevent plasma protein denaturation. The overcooling protection is forced frequency reduction when the box temperature is less than -50℃.

[0015] As a further improvement of the present technical solution, in S8, historical data is used to establish a refrigeration efficiency characteristic library, and the optimal PID parameter combination is obtained according to the load, load temperature and set temperature. The sliding window regression prediction is used, and the corresponding formula is: ; ; ; in, 、 and They are The proportional coefficient, integral coefficient and differential coefficient at the moment, 、 and They are The proportional coefficient, integral coefficient and differential coefficient at the moment, 、 and They are The proportional coefficient, integral coefficient and differential coefficient at the moment, is the load change sensitivity coefficient, is the load variation.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. The intelligent control method for this multi-refrigeration unit plasma quick freezer uses a temperature protection threshold setting to avoid the risk of plasma protein denaturation caused by rapid cooling or overcooling. Combined with the high-precision temperature tracking of the fuzzy PID control algorithm, it ensures that the biological activity of plasma is not damaged during the quick freezing process, meeting the stringent quality requirements of medical-grade cold chains for biological products.

[0017] 2. In the intelligent control method of the plasma quick-freezing machine with multiple refrigeration units, a polling scheduling mechanism is used to dynamically allocate the working cycle of each unit to avoid long-term high-load operation of a single refrigeration unit. The speed change rate is limited to reduce wear of mechanical components. At the same time, the gradient descent method and the improved whale algorithm are used to optimize the control parameters in real time, reduce compressor current fluctuations, and extend the overall service life.

[0018] 3. In the intelligent control method of the plasma quick-freezing machine with multiple refrigeration units, the sequential delayed start-up effectively disperses the starting current of multiple units, and the start-up interval is intelligently adjusted in conjunction with the grid voltage fluctuation threshold to reduce the instantaneous impact on the grid, meeting the high requirements of medical equipment for power supply stability. When multiple refrigeration units work together, high-frequency start-stop and sudden power changes are avoided through dynamic speed distribution and working cycle optimization, thereby improving the compatibility of the system with the grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a workflow diagram of the present invention; DETAILED DESCRIPTION

[0020] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0021] like Figure 1 As shown, this embodiment provides an intelligent control method for a plasma quick-freezing machine with a multi-refrigeration unit combination, comprising the following steps: S1: After power-on, several refrigeration units are started sequentially and delayed to avoid the impact of high-power starting current on the power grid. After starting, the refrigeration units are operated at full power to quickly reduce the temperature of the upper and lower cold plates of the quick freezer; S2. Optimize the refrigeration unit that needs to be turned on based on the number of blood bags placed in the blood bag according to the weight sensor; S3, collecting the load quantity and load temperature data in the quick freezer in real time through the weight sensor and temperature sensor; S4. Input the load amount and load temperature data into the control unit and generate a speed adjustment command using a PID control algorithm based on fuzzy logic. The algorithm specifically includes the steps of loading preset parameters to initialize PID parameters and fuzzy rule weights, collecting temperature error and change rate, fuzzification processing, establishing and matching fuzzy rules, defuzzification calculation, outputting the speed adjustment value and performing integral separation and anti-saturation; S5. When the load increases or the temperature rises, the control unit increases the speed of the refrigeration unit; when the load decreases or the temperature approaches a set threshold, the control unit decreases the speed of the refrigeration unit; S6. Based on the speed adjustment result, dynamically allocate the working cycle and start-stop sequence of each refrigeration unit; S7, through the closed-loop feedback mechanism, dynamically correct the speed adjustment parameters according to the actual cooling effect; S8. Use historical data to establish a refrigeration efficiency feature library, give priority to starting the unit with the highest energy efficiency ratio, obtain the optimal PID parameter combination based on load, load temperature and set temperature, and use sliding window regression prediction.

[0022] In this embodiment, the delayed start time in S1 is 3-10 seconds, and the specific duration is determined by the grid voltage fluctuation threshold. This setting can effectively disperse the starting current of multiple units and reduce the instantaneous impact on the grid.

[0023] In this embodiment, in S4, the load amount and load temperature data are input into the control unit, and a PID control algorithm based on fuzzy logic is used to generate a speed adjustment instruction. The specific steps are as follows: S4.1, load preset parameters, initialize PID parameters and fuzzy rule weights, all rules are equally important, the user enters the set temperature ; S4.2, data collection, temperature error and temperature error rate of change As an input variable, the temperature error To set the temperature With actual temperature The difference, set temperature Set by the user, actual temperature The data measured by the temperature sensor, the temperature error change rate is the temperature error The differential value of , the corresponding formula is: ; ; Among them, the temperature error The domain of , temperature error change rate The domain of ; S4.3, fuzzy processing, the temperature error and temperature error rate of change Map to fuzzy sets, calculate its membership in each fuzzy set, and use triangular function to divide seven fuzzy sets, which are {NB, NM, NS, ZO, PS, PM, PB}, corresponding to {negative large, negative medium, negative small, zero, positive small, positive medium, positive large}; S4.4. Establish 49 rules (7×7 matrix) based on expert opinions. Example rules are as follows: like and ,but , meet the demand for rapid cooling and greatly increase the speed; like and ,but , maintain a stable state; like and ,but , fine-tune the performance and slightly reduce the speed; S4.5, matching fuzzy rules, according to the fuzzy set combination of input variables, find the corresponding output fuzzy set from the fuzzy rule library to obtain the speed adjustment gradient , speed adjustment gradient The domain of ; S4.6, defuzzification calculation, using the centroid method to convert the fuzzy output into an exact value, the corresponding formula is: ; in, To adjust the speed gradient accurately, For the The membership of the fuzzy set is output by the rules. For the The output value corresponding to the rule; S4.7, output refrigeration unit speed adjustment value, the corresponding formula is: ; in, is the speed adjustment value of the refrigeration unit; S4.8, integral separation anti-saturation, add integral separation strategy, if , freeze the integral term , to prevent speed overshoot.

[0024] In this embodiment, in S6, the speed adjustment amount of the refrigeration unit is adjusted according to the speed of the refrigeration unit. Allocate work cycles to each unit , the corresponding formula is: ; in, is the maximum permissible speed of the refrigeration unit; The refrigeration unit adopts polling scheduling to avoid a single unit running at high load for a long time.

[0025] In this embodiment, in S7, the speed adjustment parameter is dynamically corrected according to the actual cooling effect through a closed-loop feedback mechanism. The specific steps include: S7.1. Adjusting fuzzy rule weights , minimize the temperature error , update the fuzzy rule weights according to the gradient descent method, and the weight update direction is in the opposite direction of the error gradient, gradually reducing the temperature deviation, so that the actual cooling process is closer to the theoretical expectation. Through continuous iterative optimization, the fuzzy rule has an effect on the temperature error change rate. and temperature error The response is more accurate, avoiding control lag or overshoot caused by fixed rule weights, ensuring strict temperature control during plasma quick freezing and protecting biological activity. The corresponding formula is: ; in, is the updated fuzzy rule weight, is the fuzzy rule weight before updating, is the learning rate, is the theoretical cooling rate, is the actual cooling rate, is the partial derivative of the speed regulation of the refrigeration unit with respect to the fuzzy rule weight, Through chain rule calculation, the weight of rules can also be restricted to prevent the rules from being over-strengthened or weakened; The corresponding multi-objective optimization function is: ; in, is a multi-objective optimization function, 、 and are weight coefficients, is the absolute value of temperature deviation, is the refrigeration unit power, To ensure system stabilization time, it can flexibly balance medical-grade temperature control accuracy and equipment energy efficiency, avoiding energy waste or stability degradation caused by over-optimization of a single goal; S7.2. Proportional coefficient based on improved whale algorithm , integral coefficient and differential coefficients The specific steps for optimization are: A1. Initialize the parameters, set the population size, spatial dimension to three, maximum number of iterations and initial population of whales, the position vector of the whale is ( , , ); A2. Calculate the fitness value. Calculate the fitness value according to the fitness function. The fitness function is: ; in, is the fitness function, is the time-weighted integrated absolute error, is the running time; A3. Boundary condition processing: process each individual in the population and calculate the coefficient vector and , and generate uniformly distributed decision random numbers , the corresponding formula is: ; ; ; in, Used to control the strength of the whale's attraction or repulsion towards or away from a target, Used to calculate the distance vector, The size and direction of the impact distance, is a random vector on [0,1], is the control vector, is the initial value of the control vector, is the current iteration number, The maximum number of iterations enables the algorithm to dynamically adjust the search granularity of the PID parameters according to the iteration progress, avoiding blind search in the early stage and oscillation in the later stage; A4. Randomly search for prey and introduce Lecy flight: and When whales randomly wander around the global space for food, their survey range is relatively wide. The introduction of Lecy flight enables individual whales to jump out of the current local area and explore in a wider solution space. The traditional whale algorithm relies on a fixed step size and is prone to falling into local optimality. The long-distance jumping characteristics of Lecy flight (the step size follows a power-law distribution) can effectively cover unexplored areas and increase population diversity. It is especially suitable for global optimization in high-dimensional parameter space. The corresponding formula is: ; in, is the position vector of a randomly selected individual in the current population, For the current whale The position in the iteration, For the current whale The position in the iteration, In order to obey the random step size, in the early stage of the algorithm, Lecy flight helps to quickly scan the entire parameter range, avoiding the search blind spots caused by the uneven distribution of the initial population, and laying the foundation for subsequent local fine optimization; A5. Shrink and surround the prey, introduce suboptimal and third-optimal solutions: and When , it means that the humpback whale has found its prey and is shrinking and encircling it. At this time, the current global optimal whale position is used as the target prey position. The whale updates its position based on the current global optimal whale position, considers the suboptimal solution and the third optimal solution, avoids the deviation of the single optimal solution (such as the pseudo-optimal solution caused by accidental noise), and uses the neighborhood information of the suboptimal solution to correct the search direction. The corresponding formula is: ; in, 、 and Respectively The optimal, suboptimal, and third-optimal position vectors of the objective function in the swarm in the iterations. Multiple solutions collaboratively guide the whales to gather in more reliable optimal areas. Especially when there are multiple high-fitness areas in the solution space, the "guidance" of suboptimal solutions can prevent the algorithm from converging to local suboptimal solutions too early. A6. Spiral update position: When When , it means that the whale is performing spiral bubble hunting. First, the distance between the current whale and the current optimal position whale is calculated, and then a spiral equation is established between the positions of the two. That is, the whale is hunting prey with a spiral bubble net. is the distance between the current whale and the current optimal whale position, and the corresponding formula is: ; ; in, are the coefficients used to define the shape of the logarithmic spiral, is a uniformly distributed random number between [-1,1]. is the coefficient The initial value of is the decay rate parameter. Through the smooth decay of the cosine function, the adaptive switching of the search strategy from "global exploration" to "local development" is achieved, taking into account the search breadth and depth of the algorithm. A7. Determine the termination condition. Stop the iteration when the maximum number of iterations is reached. Otherwise, repeat steps A3 to A6.

[0026] Furthermore, in S7, the trigger conditions for dynamic correction include periodic correction and event-driven correction. The periodic correction is to perform parameter fine-tuning every 30 seconds. The event-driven correction includes the change in the number of blood bags exceeding 2 bags, the temperature difference of the box exceeding 0.5℃, and the compressor current fluctuation in the refrigeration unit exceeding 15%, to avoid temperature control failure caused by sudden load changes or equipment performance degradation. In S7, a safety constraint mechanism is also introduced, which includes speed change rate limit and temperature protection threshold. The speed change rate limit is ,in, The temperature protection threshold includes the maximum cooling rate limit and overcooling protection. The maximum cooling rate limit is less than or equal to 3℃ / min to prevent plasma protein denaturation. The overcooling protection is forced frequency reduction when the box temperature is less than -50℃ to ensure safe and stable operation of the system.

[0027] In this embodiment, in S8, historical data is used to establish a cooling efficiency feature library, and the optimal PID parameter combination is obtained based on the load, load temperature, and set temperature. Sliding window regression prediction is used to improve the system's adaptability to different load conditions. The corresponding formula is: ; ; ; in, 、 and They are The proportional coefficient, integral coefficient and differential coefficient at the moment, 、 and They are The proportional coefficient, integral coefficient and differential coefficient at the moment, 、 and They are The proportional coefficient, integral coefficient and differential coefficient at the moment, is the load change sensitivity coefficient, is the load variation.

[0028] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely preferred examples of the present invention and are not intended to limit the present invention. Various changes and improvements may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and improvements fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent control method for a plasma quick-freezing machine with multiple refrigeration units, characterized in that: The steps include: S1: After power-on, several refrigeration units are started sequentially and delayed, and then the refrigeration units are operated at full power; S2. Optimize the refrigeration unit that needs to be turned on based on the number of blood bags placed in the blood bag according to the weight sensor; S3, collecting the load quantity and load temperature data in the quick freezer in real time through the weight sensor and temperature sensor; S4, inputting the load amount and load temperature data into the control unit, and using the PID control algorithm based on fuzzy logic to generate a speed adjustment instruction; S5. When the load increases or the temperature rises, the control unit increases the speed of the refrigeration unit; when the load decreases or the temperature approaches a set threshold, the control unit decreases the speed of the refrigeration unit; S6. Based on the speed adjustment result, dynamically allocate the working cycle and start-stop sequence of each refrigeration unit; S7, through the closed-loop feedback mechanism, dynamically correct the speed adjustment parameters according to the actual cooling effect; S8. Use historical data to establish a refrigeration efficiency feature library.

2. The intelligent control method for a plasma quick-freezing machine with multiple refrigeration units according to claim 1, characterized in that: The delay start time in S1 is 3-10 seconds, and the specific duration is determined by the grid voltage fluctuation threshold.

3. The intelligent control method for a plasma quick-freezing machine with multiple refrigeration units according to claim 1 is characterized in that: In S4, the load amount and load temperature data are input into the control unit, and a PID control algorithm based on fuzzy logic is used to generate a speed adjustment instruction. The specific steps are as follows: S4.1, load preset parameters, initialize PID parameters and fuzzy rule weights, and user inputs set temperature ; S4.2, data collection, temperature error and temperature error rate of change as input variables; S4.3, fuzzy processing, the temperature error and temperature error rate of change Map to fuzzy sets and calculate its membership in each fuzzy set; S4.

4. Establish rules based on expert opinions; S4.5, matching fuzzy rules, according to the fuzzy set combination of input variables, searching the corresponding output fuzzy set from the fuzzy rule base; S4.6, defuzzification calculation, using the centroid method to convert the fuzzy output into an exact value; S4.7, output refrigeration unit speed adjustment value; S4.8, integral separation anti-saturation, add integral separation strategy.

4. The intelligent control method for a plasma quick-freezing machine with multiple refrigeration units according to claim 1 is characterized in that: In the step S6, the speed of the refrigeration unit is adjusted according to the speed of the refrigeration unit. Allocate work cycles to each unit , the refrigeration unit adopts polling scheduling.

5. The intelligent control method for a plasma quick-freezing machine with multiple refrigeration units according to claim 1 is characterized in that: In S7, the speed adjustment parameter is dynamically corrected according to the actual cooling effect through a closed-loop feedback mechanism. The specific steps include: S7.

1. Update the fuzzy rule weights according to the gradient descent method; S7.

2. Proportional coefficient based on improved whale algorithm , integral coefficient and differential coefficients Perform optimization.

6. The intelligent control method for a plasma quick-freezing machine with multiple refrigeration units according to claim 5, characterized in that: In S7.1, the fuzzy rule weights are updated according to the gradient descent method, and the corresponding formula is: ； in, is the updated fuzzy rule weight, is the fuzzy rule weight before updating, is the learning rate, is the theoretical cooling rate, is the actual cooling rate, is the partial derivative of the speed regulation of the refrigeration unit with respect to the fuzzy rule weight.

7. The intelligent control method for a plasma quick-freezing machine with multiple refrigeration units according to claim 5, characterized in that: In S7.1, the fuzzy rule weights are updated according to the gradient descent method, and the corresponding multi-objective optimization function is: ； in, is a multi-objective optimization function, 、 and are weight coefficients, is the absolute value of temperature deviation, is the refrigeration unit power, It is the system stabilization time.

8. The intelligent control method for a plasma quick-freezing machine with multiple refrigeration units according to claim 5, characterized in that: In S7.2, based on the improved whale algorithm, the proportional coefficient , integral coefficient and differential coefficients The specific steps for optimization are: A1. Initialize parameters, set the population size, spatial dimension to three, maximum number of iterations, and initial whale population; A2. Calculate the fitness value according to the fitness function; A3. Boundary condition processing; A4, random search for prey, introduction of Lecy flight; A5. Shrink and surround the prey, introducing suboptimal and third-optimal solutions; A6, spiral update position; A7. Determine the termination condition. Stop the iteration when the maximum number of iterations is reached. Otherwise, repeat steps A3 to A6.

9. The intelligent control method for a plasma quick-freezing machine with multiple refrigeration units according to claim 1, characterized in that: In S7, the trigger conditions for dynamic correction include periodic correction and event-driven correction. The periodic correction is to perform parameter fine-tuning every 30 seconds. The event-driven correction includes the change in the number of blood bags exceeding 2 bags, the temperature difference of the box exceeding 0.5°C, and the current fluctuation of the compressor in the refrigeration unit exceeding 15%. In S7, a safety constraint mechanism is also introduced. The safety constraint mechanism includes speed change rate limit and temperature protection threshold. The speed change rate limit is ,in, The temperature protection threshold includes the maximum cooling rate limit and overcooling protection. The maximum cooling rate limit is less than or equal to 3℃ / min to prevent plasma protein denaturation. The overcooling protection is forced frequency reduction when the box temperature is less than -50℃.

10. The intelligent control method for a plasma quick-freezing machine with multiple refrigeration units according to claim 1, characterized in that: In S8, historical data is used to establish a refrigeration efficiency feature library, and the optimal PID parameter combination is obtained according to the load amount, load temperature and set temperature, and a sliding window regression prediction is adopted.

Images