An intelligent control method and system for cleaning a raw material bin

Abstract

The present invention relates to the technical field of data processing, and particularly relates to an intelligent control method and system for cleaning a raw material bin. The method includes: obtaining in real time the flow rate data and pressure data of the cleaning agent during the current time period when cleaning the raw material bin; calculating the abnormality degree of the flow rate data in the previous time period; calculating the intensity degree of the flow rate data in the previous time period; determining the adaptive cooling coefficient of the flow rate data in the current time period; and based on the adaptive cooling coefficient, using the simulated annealing algorithm to obtain the ideal flow rate of the flow rate data in the next time period, so as to realize the intelligent control of the cleaning of the raw material bin. By dynamically adjusting the cooling coefficient according to the abnormality degree and intensity degree of the flow rate data, the present invention improves the robustness and adaptability of the simulated annealing algorithm, overcomes the limitations brought by the fixed cooling coefficient; monitors and adjusts the flow rate in real time to ensure the stability of the cleaning agent flow rate, improves the quality of the cleaning of the raw material bin, and avoids the influence of errors or fluctuations on the cleaning effect.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to an intelligent control method and system for cleaning a raw material bin. Background Art

[0002] With the increasingly high requirements of modern industrial enterprises for the management and cleaning of raw material warehouses, the cleaning of raw material bins is not only related to production quality, but also involves the safety and hygiene standards of products. Traditional manual cleaning methods are inefficient and prone to human errors, which do not meet the requirements of modern production enterprises for high efficiency, safety, and error-free. With the development of technology, intelligent cleaning systems can reduce manual intervention through automatic control, improve cleaning efficiency, shorten the cleaning cycle, and thus effectively improve the overall efficiency of the production line. In an intelligent cleaning system, the intelligent control of the cleaning agent flow directly affects the cleaning effect; if the cleaning agent flow is insufficient, it may not be able to effectively remove the residues or dirt in the raw material bin, resulting in incomplete cleaning; if the cleaning agent flow is too large, it will cause waste, not only increasing the cleaning cost, but also possibly causing unnecessary corrosion or damage to the raw material bin. There is currently a method for intelligent control of cleaning agent flow, which is the simulated annealing algorithm. This algorithm has strong flexibility and adaptability in the intelligent control of cleaning agent flow and has good global optimization ability.

[0003] The patent document with the publication number CN112232545B discloses an AGV task scheduling method based on the simulated annealing algorithm, belonging to the field of task scheduling in an automated warehousing system. It includes: 1. determining the position coordinates and cruising range of the AGV, and the position coordinates of the shelves to be transported and the picking workstations; 2. setting the parameters of the simulated annealing algorithm; 3. generating an initial task allocation plan; 4. starting iterative search to generate candidate solutions; 5. checking the feasibility of the candidate solutions, and if they are not feasible, turning to step 4 to regenerate candidate solutions; 6. judging whether to accept the candidate solutions, and if accepted, replacing the current solution with the candidate solutions; 7. if the preset number of iterations is not reached, jumping to step 4 for the next iterative search; 8. if the algorithm termination condition is met, outputting the current solution, otherwise reducing the annealing temperature and jumping to step 4 to start a new round of iterative search. This method can quickly solve the AGV task scheduling problem in an intelligent warehousing system environment and obtain a task allocation plan with the shortest or approximately shortest completion time.

[0004] However, the above patent documents are not directed at the direction of warehouse cleaning. When using the simulated annealing algorithm for intelligent control of the cleaning agent flow data, a fixed cooling coefficient (i.e., the temperature decrease rate) is usually used, which affects the exploration and convergence process of the algorithm. If the cooling coefficient is too small, the convergence speed will be too slow, thus wasting a large amount of computing resources. If the cooling coefficient is too large, the temperature will drop too fast, and the algorithm will quickly stop exploring new solutions and prematurely converge to a local optimal solution. In this way, although the algorithm runs fast, the final result may not be the global optimal solution. At the same time, when collecting the cleaning agent flow data, due to the possible electromagnetic interference from other electrical equipment (such as motors, frequency converters, etc.) in the cleaning system, noise data is generated in the cleaning agent flow data. Then the existence of the noise data will affect the quantification of the change characteristics of the cleaning agent flow data and the accuracy of the subsequent calculation of the cooling coefficient. Summary of the Invention

[0005] In order to solve the problem that using a fixed cooling coefficient in the simulated annealing algorithm affects the exploration and convergence process of the algorithm, and at the same time, due to the noise data generated by electromagnetic interference, the accuracy of calculating the cooling coefficient is reduced, the present invention provides an intelligent control method and system for raw material warehouse cleaning.

[0006] In the first aspect, the present invention provides an intelligent control method for raw material warehouse cleaning, adopting the following technical solutions:

[0007] An intelligent control method for raw material warehouse cleaning includes: acquiring in real time the flow data and pressure data of the cleaning agent during the current time period when cleaning the raw material warehouse; calculating the abnormality degree of the flow data in the previous time period based on the range of the flow data points in the flow data of the previous time period and the number of flow data points with a second derivative of 0 in the flow data of the previous time period; calculating the severity degree of the flow data in the previous time period based on the abnormality degree, the correlation coefficient between the flow data and pressure data in the previous time period, and the difference between the percentiles of the flow data points and pressure data points corresponding to all moments in the previous time period in their respective data segments; weighting the preset initial cooling coefficient according to the severity degree to obtain the adaptive cooling coefficient of the flow data in the current time period; and obtaining the ideal flow of the flow data in the next time period by using the simulated annealing algorithm according to the adaptive cooling coefficient to realize the intelligent control of raw material warehouse cleaning.

[0008] By dynamically adjusting the cooling coefficient according to the abnormality degree and severity of the flow data, the robustness and adaptability of the simulated annealing algorithm are improved, and the limitations brought by the fixed cooling coefficient are overcome; the flow is monitored and adjusted in real time to ensure the stability of the cleaning agent flow, improve the quality of the raw material bin cleaning, and avoid the influence of errors or fluctuations on the cleaning effect; by calculating the outliers of the flow and pressure data, the influence of electromagnetic interference and noise on the data is reduced, and the accuracy of the algorithm optimization is ensured; the flow is precisely controlled to reduce resource waste, optimize the cleaning process, and improve production efficiency.

[0009] Further, the abnormality degree satisfies the following relational expression:

[0010] ; where is the abnormality degree of the flow data in the previous time period, and are respectively the maximum and minimum values of the flow data points in the flow data of the previous time period, is the number of flow data points with a second derivative of 0 in the flow data of the previous time period, is a hyperparameter, is a normalization function, is the absolute value symbol.

[0011] By combining the range of the flow data and the number of flow data points with a second derivative of 0, the abnormal fluctuations in the flow data can be accurately identified, the abnormality degree can be quantified, the sensitivity to data changes can be improved, and the smoothness of the data can be reflected by the number of flow data points with a second derivative of 0, effectively reducing the influence of noise on the flow data and improving the accuracy of data analysis; through normalization processing and the introduction of hyperparameters, the calculation of the abnormality degree is more stable for data of different scales, ensuring better adaptability of the cooling coefficient adjustment.

[0012] Further, the severity satisfies the following relational expression:

[0013] ; where is the severity of the flow data in the previous time period, is the abnormality degree of the flow data in the previous time period, is the correlation coefficient between the flow data and the pressure data in the previous time period, is the number of time instants in the previous time period, is the percentile of the flow data point corresponding to the -th time instant in the previous time period in the corresponding data segment, is the percentile of the pressure data point corresponding to the -th time instant in the previous time period in the corresponding data segment, is a hyperparameter, is a normalization function, is the absolute value symbol.

[0014] By combining the abnormality degree of the flow data and the correlation coefficient between the flow and pressure data, the drastic change of the flow data can be accurately measured, reflecting the severity of the fluctuations in the system; introducing the percentile difference of the flow and pressure data points provides a more comprehensive evaluation, which can consider the deviation between the flow and pressure simultaneously, making the calculation more accurate.

[0015] Further, the correlation coefficient is the Pearson correlation coefficient.

[0016] Further, the adaptive cooling coefficient satisfies the following relational expression:

[0017] ; where, is the adaptive cooling coefficient of the flow data in the current time period, is the severity of the flow data in the previous time period, is the preset initial cooling coefficient.

[0018] By combining the severity of the flow data in the previous time period with the preset initial cooling coefficient, the cooling coefficient in the current time period can be dynamically adjusted according to the strength of the flow fluctuation, ensuring that the cooling system makes an optimized adjustment according to the real-time demand and improving the cooling efficiency; the adaptive cooling coefficient can be adjusted according to the drastic change of the system flow, avoiding over-cooling or under-cooling, thus effectively maintaining the stable operation of the system.

[0019] Further, obtaining the ideal flow of the flow data in the next time period by using the simulated annealing algorithm includes: obtaining the dirt concentration data of the raw material bin in the current time period in real time, constructing the objective function of the simulated annealing algorithm based on the flow data in the current time period, the values of the dirt concentration data points corresponding to the start and end times of the raw material bin in the current time period, and the cleaning cost of the raw material bin in the current time period, and taking the optimal solution in the solution of the objective function as the ideal flow of the flow data in the next time period.

[0020] By comprehensively considering the current flow, dirt concentration and cleaning cost, the simulated annealing algorithm can accurately optimize the ideal flow in the next time period, improving the system operation efficiency; obtaining the dirt concentration data in real time, the system can adjust the flow according to the actual dirt situation, avoiding excessive accumulation and affecting production, and maintaining stable operation; optimizing the flow data can reduce unnecessary cleaning frequency, reduce the cleaning cost, and at the same time extend the service life of the equipment; by reasonably adjusting the flow, balancing various factors, improving the production efficiency and reducing resource waste, bringing long-term energy saving and cost savings.

[0021] Further, the objective function satisfies the following relational expression:

[0022] ;

[0023] In the formula, is the solution of the objective function, is the preset fouling concentration weight value, and are the values of the fouling concentration data points corresponding to the start time and the end time of the current time period respectively, is the preset cost weight value, is the cleaning cost of the raw material bin in the current time period.

[0024] In a second aspect, the present invention provides an intelligent control system for cleaning a raw material bin, adopting the following technical solution:

[0025] An intelligent control system for cleaning a raw material bin includes: a processor and a memory, and the memory stores computer program instructions, which implement the above-mentioned intelligent control method for cleaning a raw material bin when the computer program instructions are executed by the processor.

[0026] By adopting the above technical solution, the above-mentioned intelligent control method for cleaning a raw material bin is generated into a computer program and stored in the memory to be loaded and executed by the processor, so as to manufacture a terminal device according to the memory and the processor, which is convenient to use.

[0027] The present invention has the following technical effects:

[0028] Since noise data can cause misjudgment of the change in flow data by the simulated annealing algorithm, which in turn affects the adaptive adjustment process of the cooling coefficient, by analyzing the degree of change in the flow data of the previous time period, a smoother and more accurate dynamic adjustment mechanism can be provided for the current time period, effectively reducing the interference of noise, avoiding too fast or too slow temperature drop rates, making the adjustment of the cooling coefficient more accurate, thereby optimizing the exploration process and convergence process of the simulated annealing algorithm, avoiding premature convergence to a local optimal solution, and at the same time accelerating the search speed for the global optimal solution; therefore, by analyzing the change characteristics of the cleaning agent flow data and pressure data and combining the adaptive adjustment of the cooling coefficient, the interference of noise data on the intelligent control process can be effectively reduced, and the global optimization ability and convergence speed of the simulated annealing algorithm can be improved; in addition, the adaptive adjustment mechanism can flexibly adjust the cooling rate according to the actual change of the flow data, making the control process more in line with the actual working conditions, improving the intelligent control accuracy of the cleaning agent flow, ultimately optimizing the operation efficiency of the cleaning system, reducing the waste of computing resources, and ensuring the stability and accuracy of the intelligent control results. Description of the Drawings

[0029] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features, and advantages of the exemplary embodiments of the present invention will become readily understood. In the drawings, several embodiments of the present invention are shown by way of example and not limitation, and the same or corresponding reference numerals are for the same or corresponding parts.

[0030] Figure 1 It is a flowchart of a method for intelligent control of raw material bin cleaning in an embodiment of the present invention. Specific embodiments

[0031] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0032] It should be understood that when the claims, specifications, and drawings of the present invention use terms such as "first" and "second", they are only used to distinguish different objects and not to describe a specific order. The terms "comprising" and "including" used in the specifications and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0033] The present invention analyzes the variation characteristics of the flow rate data and pressure data of the cleaning agent in the same time period, obtains the severity of the change in the flow rate data in the previous time period of the current time period, and adaptively adjusts the cooling coefficient corresponding to the flow rate data of the cleaning agent in the current time period based on the severity of the change in the cleaning agent flow rate data in the previous time period. Then, the ideal flow rate for the next time period is obtained through the simulated annealing algorithm, thereby realizing the intelligent control of raw material bin cleaning and improving the control accuracy.

[0034] An embodiment of the present invention discloses a method for intelligent control of raw material bin cleaning, referring to Figure 1 , including steps S1 - S5:

[0035] S1: Real-time obtain the flow rate data and pressure data of the cleaning agent during the current time period when cleaning the raw material bin.

[0036] In the present invention, a high-frequency turbine flowmeter and a pressure sensor are used to collect the flow rate data and pressure data of the cleaning agent of the high-pressure spray gun during the cleaning process in real time. It is necessary to ensure that the two types of data are collected in the same time period. The specific collection time is the entire process of cleaning the raw material bin, and the collection frequency is once every two seconds. Then, an analog-to-digital conversion device is used to perform analog-to-digital conversion on the above two types of data.

[0037] S2: Calculate the abnormality degree of the flow data in the previous time period.

[0038] It should be noted that by analyzing the data change characteristics of the flow data of the cleaning agent, the abnormality degree of each time period is obtained. When analyzing this index, the more discrete the numerical distribution in the flow data of a time period and the more frequent the changes, the greater the abnormality degree of the flow data of a time period can be indicated.

[0039] Based on the range of the flow data points in the flow data of the previous time period and the number of flow data points with a second derivative of 0 in the flow data of the previous time period, calculate the abnormality degree of the flow data of the previous time period.

[0040] Specifically, the abnormality degree satisfies the following relational expression:

[0041] ;

[0042] In the formula, is the abnormality degree of the flow data of the previous time period, and are respectively the maximum value and the minimum value of the flow data points in the flow data of the previous time period, is the number of flow data points with a second derivative of 0 in the flow data of the previous time period, is a hyperparameter, is a normalization function, is the absolute value symbol.

[0043] Implementers can set the hyperparameter according to the specific implementation situation. For example, 0.1. The existence of the hyperparameter is to prevent when, making the formula meaningless.

[0044] Among them, represents the range of the flow data points in the flow data of the previous time period. The larger this value is, the greater the abnormality degree of the flow data of the previous time period can be indicated; The larger it is, the more frequent the numerical changes in the flow data of the previous time period can be indicated, which can confirm the greater credibility of the larger range in the flow data of the previous time period, and then the abnormality degree of the flow data of the previous time period will be greater.

[0045] S3: Calculate the intensity of the flow data in the previous time period.

[0046] It should be noted that since there may be noise data in the data, which may cause the originally normal flow data to show the same abnormal degree as the abnormal flow data, it is necessary to exclude the possibility of noise data. According to the scenario investigation, at the nozzle of the high-pressure spray gun, the pressure energy of the fluid is converted into kinetic energy, resulting in an increase in the flow velocity. Therefore, the flow rate of the spray gun and the pressure of the system are often positively correlated. If there is noise data in the flow rate data, the correlation between the two types of data will decrease. Then, in this step, the correlation between the change characteristics of the flow rate data and the pressure data of the cleaning agent will be analyzed, and the severity of the flow rate data will be obtained by combining the abnormal degree of the flow rate data. Among them, the greater the correlation between the change characteristics of the flow rate data and the pressure data of the cleaning agent, the smaller the possibility that the flow rate data in this time period is affected by noise. If the abnormal degree of the flow rate data in this time period is relatively large, the possibility of belonging to noise is smaller and the possibility of belonging to the real change of the high-pressure spray gun itself is greater. Then, the severity of the flow rate data in this time period will also be greater.

[0047] Based on the abnormal degree, the correlation coefficient of the flow rate data and the pressure data in the previous time period, and the difference between the percentiles of the flow rate data points and the pressure data points corresponding to all moments in the previous time period in their respective data segments, calculate the severity of the flow rate data in the previous time period.

[0048] Specifically, the severity satisfies the following relational expression:

[0049] ;

[0050] In the formula, is the severity of the flow rate data in the previous time period, is the abnormal degree of the flow rate data in the previous time period, is the correlation coefficient of the flow rate data and the pressure data in the previous time period, is the number of moments in the previous time period, is the th moment in the previous time period, and is the percentile of the corresponding flow rate data point in its respective data segment, is the th moment in the previous time period, and is the percentile of the corresponding pressure data point in its respective data segment, is a hyperparameter,

[0051] The implementer can set the hyperparameter according to the specific implementation situation. For example, 0.1. The existence of the hyperparameter is to prevent from making the formula meaningless when

[0052] Specifically, the correlation coefficient is the Pearson correlation coefficient.

[0053] Among them, The larger it is, the more drastic and frequent the data changes in the flow data of the previous time period are. Then, the severity of the flow data in the previous time period will also be greater. The larger it is, the greater the correlation between the flow data and the pressure data in the previous time period. Then, when the abnormal degree of the flow data in the previous time period is relatively large, the possibility that the reason for this situation belongs to noise is smaller, and the possibility that it belongs to the actual change of the high-pressure spray gun itself is greater. Then, the severity of the flow data in the previous time period will also be greater. represents the sum of the differences between the percentile values of all the flow data points and pressure data points corresponding to all moments in the previous time period in their respective data segments. The smaller this value is, the greater the correlation between the flow data and the pressure data in the previous time period can be shown, and it can also confirm the greater credibility of the Pearson correlation coefficient between the flow data and the pressure data in the previous time period. Then, when the abnormal degree of the flow data in the current time period is relatively large, the possibility that the reason for this situation belongs to noise is smaller, and the possibility that it belongs to the actual change of the high-pressure spray gun itself is greater. Then, the severity of the flow data in the previous time period will also be greater.

[0054] S4: Determine the adaptive cooling coefficient of the flow data in the current time period.

[0055] It should be noted that when making the cooling coefficient of the flow data adaptive, first, it is necessary to clarify that the regulation frequency of the flow data in the present invention is that one flow data regulation value corresponds to one time period (including multiple sampling moments), rather than one flow data regulation value corresponding to one sampling moment. This is because regulating the flow data based on time periods can effectively solve problems such as equipment instability, high control complexity, and resource waste caused by instantaneous flow fluctuations. Then, determine a time period (for example, with a time length of one minute) as a benchmark to regulate the flow data. Finally, when calculating the regulation value of the flow data corresponding to one time period, it is necessary to adapt its cooling coefficient, and the reference basis is the severity of the flow data in the previous time period immediately adjacent to this time period. (Although the flow data in the previous time period may also correspond to one flow data regulation value, the flow data will not completely remain at a constant regulation value within one time period but will fluctuate within a reasonable range. Therefore, there will be different degrees of severity for reference.) Therefore, taking the current time period as an example, based on the severity of the flow data in the previous time period of the current time period, calculate the adaptive cooling coefficient corresponding to the intelligent control of the flow data in the current time period when using the simulated annealing algorithm subsequently.

[0056] The preset initial cooling coefficient is weighted according to the severity of the flow data in the previous time period to obtain the adaptive cooling coefficient of the flow data in the current time period.

[0057] Specifically, the adaptive cooling coefficient satisfies the following relational expression:

[0058] ;

[0059] In the formula, is the adaptive cooling coefficient of the flow data in the current time period, is the severity of the flow data in the previous time period, is the preset initial cooling coefficient.

[0060] Implementers can set the initial cooling coefficient according to the specific implementation situation. For example, 0.97. If the current time period is the first time period when the raw material bin cleaning starts, the adaptive cooling coefficient in the current time period is the same as the initial cooling coefficient.

[0061] Among them, The larger it is, the larger the adaptive cooling coefficient of the flow data in the current time period should be, so that the temperature drop speed becomes slower and the number of iterations becomes more, to ensure that the iteration for the optimal solution can be more accurate.

[0062] S5: According to the adaptive cooling coefficient, use the simulated annealing algorithm to obtain the ideal flow rate of the flow data in the next time period to realize the intelligent control of the raw material bin cleaning.

[0063] Specifically, the process of using the simulated annealing algorithm to obtain the ideal flow rate of the flow data in the next time period includes:

[0064] Preset the initial parameters in the simulated annealing algorithm. For example, set the initial temperature value to the empirical value of 100 °C, the temperature threshold to the empirical value of 60 °C, the initial solution of the flow rate data of the cleaning agent to the empirical value of 20 L / min, the random perturbation range for each iteration to be within 5 L / min below the value of the flow rate setting in the previous annealing process, and the cooling coefficient to 0.97. Subsequent parameters are adaptively adjusted according to the adaptive cooling coefficient;

[0065] Obtain the dirt concentration data of the raw material bin in the current time period in real time. Based on the flow data in the current time period, the values of the dirt concentration data points corresponding to the start and end times of the raw material bin in the current time period, and the cleaning cost of the raw material bin in the current time period, construct the objective function of the simulated annealing algorithm, and then calculate and output the flow rate optimal solution corresponding to the flow data in the current time period. The specific calculation method of the optimal solution is the existing well-known technology of the simulated annealing algorithm and will not be elaborated here. Take the optimal solution in the solution of the objective function as the ideal flow rate of the flow data in the next time period.

[0066] Specifically, the objective function satisfies the following relational expression:

[0067] ;

[0068] In the formula, is the solution of the objective function, is the preset fouling concentration weight value, and are respectively the values of the fouling concentration data points corresponding to the start time and the end time of the current time period, is the preset cost weight value, is the cleaning cost of the raw material bin in the current time period.

[0069] Implementers can set the fouling concentration weight value and the cost weight value according to the specific implementation situation. For example, the fouling concentration weight value is 0.5 and the cost weight value is 0.5.

[0070] Among them, in the objective function represents the fouling removal effect within the current time period. The larger this value is, the larger the solution of the objective function is, and the worse the priority will be. In the objective function The larger it is, the larger the solution of the objective function is, and the worse the priority will be.

[0071] An embodiment of the present invention also discloses an intelligent control system for cleaning a raw material bin, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, an intelligent control method for cleaning a raw material bin according to the present invention is implemented.

[0072] The above system further includes other components well known to those skilled in the art such as a communication bus and a communication interface. Their settings and functions are known in the art, so they will not be elaborated here.

[0073] In the present invention, the aforementioned memory can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, device, or device. For example, a computer-readable storage medium can be any suitable magnetic storage medium or magneto-optical storage medium, or any other medium that can be used to store the required information and can be accessed by an application program, module, or both. Any such computer storage medium can be part of the device or accessible or connectable to the device.

[0074] Although this specification has shown and described multiple embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will envision many changes, alterations, and alternative ways without departing from the spirit and concept of the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein may be employed in practicing the present invention.

[0075] The above are all preferred embodiments of the present invention, and the protection scope of the present invention is not limited thereby. Therefore, all equivalent changes made according to the structure, shape, and principle of the present invention shall be covered within the protection scope of the present invention.

Claims

1. A raw material warehouse cleaning intelligent control method, characterized in that: include: Real-time acquisition of the flow data and pressure data of the cleaning agent in the current time period when the raw material warehouse is cleaned; Based on the range of the flow data points in the flow data of the previous time period and the number of flow data points whose second-order derivative is 0 in the flow data of the previous time period, the abnormality of the flow data of the previous time period is calculated; Calculate the severity of the flow data in the previous time period based on the abnormality, the correlation coefficient of the flow data and the pressure data in the previous time period, and the difference between the percentiles of the flow data points and the pressure data points corresponding to all moments in the previous time period in the corresponding data segments; The severity satisfies the following relationship: ； In the formula, is the intensity of the traffic data in the previous time period, is the abnormality of the traffic data in the previous time period, is the correlation coefficient between the flow data and pressure data in the previous time period, is the number of moments in the previous time period, For the previous period The percentile of the traffic data point corresponding to the moment in the data segment to which it belongs, For the previous period The percentile of the pressure data point corresponding to the moment in the data segment to which it belongs, is a hyperparameter, is the normalization function, is the absolute value symbol; The preset initial cooling coefficient is weighted according to the severity to obtain an adaptive cooling coefficient of the flow data of the current time period; According to the adaptive cooling coefficient, the ideal flow rate of the flow data for the next time period is obtained by using a simulated annealing algorithm, including: real-time acquisition of the dirt concentration data of the raw material warehouse in the current time period, and construction of an objective function of the simulated annealing algorithm based on the flow data of the current time period, the values ​​of the dirt concentration data points of the raw material warehouse corresponding to the start and end times of the current time period, and the cleaning cost of the raw material warehouse in the current time period. The optimal solution among the solutions of the objective function is used as the ideal flow rate of the flow data for the next time period, so as to realize intelligent control of the cleaning of the raw material warehouse.

2. The intelligent control method for cleaning a raw material warehouse according to claim 1, characterized in that: The abnormal degree satisfies the following relationship: ； In the formula, is the abnormality of the traffic data in the previous time period, and are the maximum and minimum values ​​of the traffic data points in the traffic data of the previous time period, is the number of flow data points whose second-order derivative is 0 in the flow data of the previous time period, is a hyperparameter, is the normalization function, is the absolute value symbol.

3. The intelligent control method for cleaning raw material warehouse according to claim 1, characterized in that: The correlation coefficient is the Pearson correlation coefficient.

4. The intelligent control method for cleaning raw material warehouse according to claim 1, characterized in that: The adaptive cooling coefficient satisfies the following relationship: ； In the formula, is the adaptive cooling coefficient of the flow data in the current time period, is the intensity of the traffic data in the previous time period, is the preset initial cooling coefficient.

5. The intelligent control method for cleaning raw material warehouse according to claim 1, characterized in that: The objective function satisfies the following relationship: ； In the formula, is the solution of the objective function, is the preset dirt concentration weight value, and are the values ​​of the dirt concentration data points corresponding to the start and end times of the current time period, is the preset cost weight value, It is the cleaning cost of the raw material warehouse in the current time period.

6. An intelligent control system for cleaning raw material warehouse, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, an intelligent control method for cleaning a raw material warehouse according to any one of claims 1 to 5 is implemented.

Images