Optimized scheduling method of air compressor system, medium and equipment

Abstract

The invention relates to an optimal scheduling method, a medium and equipment for an air compressor system. The method comprises the following steps: acquiring an air flow value of each air compressor in a certain operation period, acquiring parameter data Yn related to the air flow, calculating the correlation between the data X and the Yn, and selecting the parameter data with strong correlation; s200, the parameter data with the high correlation are predicted through a prediction model, the predicted air flow value of each air compressor is obtained, and the predicted power of each air compressor is calculated according to the predicted air flow value; and calculating a similarity index according to the predicted power and the load power of each air compressor, and when the similarity index is a positive value, determining the air compressor capable of meeting the load use requirement. According to the method, the improved prediction algorithm is adopted, the air compressor most matched with the load demand is selected by calculating the similarity index, accurate prediction is conducted according to the operation mechanism and historical data of the air compressor, and the accuracy and reliability of the prediction algorithm are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of air compressor control, and in particular to an optimization scheduling method, medium and equipment for an air compressor system. Background Art

[0002] An air compressor is a device that converts the mechanical energy of a prime mover (such as an electric motor or an internal combustion engine) into gas pressure energy. It is widely used in various fields such as industrial production, mining, construction, textiles, food, and medical treatment. With the development of industry and the need to save energy, the energy efficiency and automatic control of air compressors have become a hot topic of research. The existing air compressor control system mainly adjusts the operating status of the air compressor by detecting the operating parameters of the air compressor (such as pressure, temperature, current, etc.) to meet production needs. At the same time, some advanced control systems also use prediction algorithms to predict future air compressor operating parameters by learning from historical data, thereby adjusting the operating status of the air compressor in advance and improving energy efficiency.

[0003] However, in the existing air compressor control system, the accuracy and reliability of the prediction algorithm are still the main factors restricting its performance. The existing prediction algorithm is mainly based on the statistical analysis of historical data, lacking an in-depth understanding and consideration of the operating mechanism of the air compressor. Therefore, the accuracy and reliability of the prediction results are often poor, and it is difficult to meet the needs of actual production. In addition, when selecting air compressors, the existing air compressor control system mainly matches the parameters of the air compressor (such as flow, pressure, power, etc.), lacking comprehensive consideration of the operating status and load requirements of the air compressor. Therefore, in the actual operation process, it is easy to have excess or insufficient air compressor capacity, resulting in energy waste or low production efficiency. Therefore, how to improve the accuracy and reliability of the air compressor prediction algorithm, and how to reasonably select and dispatch air compressors to meet production needs and reduce energy consumption are problems that need to be solved in the current air compressor technology field. Summary of the invention

[0004] The object of the present invention is to provide an optimized scheduling method, medium and equipment for an air compressor system to improve the accuracy and reliability of an air compressor prediction algorithm.

[0005] To achieve the above object, the present invention provides the following technical solutions: An optimization scheduling method for an air compressor system comprises the following steps: S100, obtain the air flow value of each air compressor in a certain operation cycle, and the data is X{x1, x2, x3…x n}, and also obtain the parameter data Y related to the air flow n Where n=1, 2, 3…, the specific data of the parameter data is Y n {Y n1 , Yn2 , Y n3 …Y nn}, calculate the data X{x1, x2, x3…x n} and Y n {Y n1 , Y n2 , Y n3 …Y nn}, select parameter data with strong correlation; S200, the parameter data with strong correlation is predicted by the POA-LSTM prediction model to obtain the predicted air flow value of each air compressor, and the predicted power of each air compressor is calculated according to the predicted air flow value; S300, calculating a similarity index based on the predicted power and load power of each air compressor, and determining an air compressor that can meet the load usage requirement when the similarity index is a positive value.

[0006] The present invention further arranges that the parameter data Y related to the air flow in step S100 n Including compressed air leakage, exhaust pressure, air compressor operating current, air tank pressure, and air compressor exhaust temperature.

[0007] The present invention further provides that the air flow value X is related to the parameter data Y of the air flow n The correlation coefficient calculation formula is: ; in The value range of is [0, 1.0], the range of strong correlation is [0.9-1.0], the range of moderate correlation is [0.6-0.9), the range of medium correlation is [0.4-0.6), and the range of no correlation is [0 -0.4).

[0008] The present invention is further arranged that after obtaining the predicted air flow value, the predicted power of each air compressor is calculated according to the formula: P = k×Q×s / 60; wherein P represents the output power of the air compressor, in kilowatts; k is a constant, whose value is 1.471; Q represents the air flow, in cubic meters per minute; s is the rotation speed of the air compressor, in revolutions per minute.

[0009] The present invention further arranges that the similarity index of the predicted power and load power calculation of each air compressor is calculated as follows: Determine the operation period T, and the predicted power of each air compressor in the operation period is recorded as P N {P N1 , P N2 …, P Ni}, where P Ni is the predicted power of the air compressor in period i, and the load power in one operating cycle is PL {P L1 , P L2 …, P Li}, then the similarity index between the two is R(P N , P L ) is calculated as follows: ; ; ; ; In the above formula, is the covariance between the predicted power and load power of each air compressor; and is the variance of the predicted power and load power of each air compressor, It is the average value of the predicted power of the air compressor in one operating cycle; It is the average value of load power in one operating cycle.

[0010] The present invention is further configured to determine the air compressors that can meet the load usage requirements according to the similarity index, the number of which is Z, and establish an air compressor system energy consumption mathematical model, whose objective function is: ; In the above formula, the operating energy consumption of the EC air compressor unit; is a binary variable, representing the operating status of the jth air compressor in the time period; is the predicted air flow value of the j-th air compressor in period t; The operating energy consumption of the j-th air compressor in period t; ω is the operating energy consumption of the j-th air compressor from shutdown to startup; is the air compressor running time; Z is the air compressor that can meet the load usage requirements.

[0011] The present invention is further arranged, wherein The relationship with the predicted air flow value is as follows: ; a j 、b j 、c j ——Fitting parameters of the energy consumption characteristic curve of the jth air compressor.

[0012] The present invention is further arranged to establish constraint conditions during scheduling, and the constraint conditions include load balance constraint, gas production constraint, pipeline pressure constraint, air compressor start and stop times constraint and binary variable constraint.

[0013] The present invention also provides a computer device, which includes a memory and a processor. The memory is used to store computer instructions, and the processor executes the optimization scheduling method of the air compressor system by executing the computer instructions.

[0014] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the optimization scheduling method for an air compressor system.

[0015] Beneficial effects of the present invention: Compared with the existing technology, the present invention includes at least one of the following beneficial effects: 1. Improve prediction accuracy: The present invention adopts an improved prediction algorithm, selects the air compressor that best matches the load demand by calculating the similarity index, and makes accurate predictions based on the operating mechanism and historical data of the air compressor, thereby improving the accuracy and reliability of the prediction algorithm. 2. Optimize resource allocation: When selecting an air compressor, the present invention not only considers the parameters of the air compressor, but also comprehensively considers the operating status and load demand of the air compressor, thereby realizing the reasonable selection and scheduling of the air compressor, avoiding the situation of excess or insufficient capacity of the air compressor, optimizing resource allocation, and improving production efficiency. 3. Reduce energy consumption: Under the premise of ensuring the load use requirements, the present invention minimizes the operating energy consumption of the air compressor unit, and realizes efficient use of energy and reduces energy consumption by optimizing the selection and scheduling of the air compressor. 4. Improve system energy efficiency: The energy consumption mathematical model of the air compressor system of the present invention takes into account the energy consumption characteristic curve and operating status of the air compressor, and realizes the improvement of the energy efficiency of the air compressor system by optimizing the start and stop and operating status of the air compressor. 5. High flexibility: The scheduling method of the present invention can flexibly adjust the selection and scheduling strategy of the air compressor according to different load requirements and environmental conditions, thereby improving the adaptability and flexibility of the system. In summary, compared with the prior art, the present invention has higher prediction accuracy, better resource allocation, lower energy consumption, higher system energy efficiency and higher flexibility. It is an advanced air compressor control method with important practical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings: Figure 1 Schematic diagram of a flow chart of an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The following will describe the implementation methods of the present application in detail with the help of accompanying drawings and examples, so that the implementation process of how the present application uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly.

[0018] Embodiment 1: like Figure 1 As shown, this embodiment provides an optimization scheduling method for an air compressor system, comprising the following steps: S100, obtain the air flow value of each air compressor in a certain operation cycle, and the data is X{x1, x2, x3…x n}, where certain operation settings are 0 o'clock-24 o'clock; and obtain parameter data Y related to air flow n Where n=1, 2, 3…, the specific data of the parameter data is Y n {Y n1 , Y n2 , Y n3 …Y nn}, calculate the data X{x1, x2, x3…x n} and Y n {Y n1 , Y n2 , Y n3 …Y nn}, select the parameter data with strong correlation; parameter data Y related to air flow n Including compressed air leakage, exhaust pressure, air compressor operating current, gas tank pressure, and air compressor exhaust temperature. The parameter data with strong correlation used in this embodiment are: air compressor exhaust temperature, air compressor operating current, exhaust pressure, and gas tank pressure. These four data are used as input values ​​of the POA-LSTM prediction model. S200, the parameter data with strong correlation is predicted by the POA-LSTM prediction model to obtain the predicted air flow value of each air compressor, and the predicted power of each air compressor is calculated according to the predicted air flow value; To establish a prediction model, the LSTM network must first be trained. The training process includes two parts: forward signal transmission and error back propagation. The values ​​of the hidden layer and the output layer are calculated based on the forward signal transmission. The connection weights are optimized by back propagating the error between the actual output value and the theoretical output value until the error is infinitely small or less than the given reference error value. The training is completed and the network model parameters are saved.

[0019] The steps of the POA-LSTM prediction model for prediction are as follows: The four data items, namely, air compressor exhaust temperature, air compressor operating current, exhaust pressure and air tank pressure, are used as input values ​​of the POA-LSTM prediction model and normalized so that the input features have the same measurement scale and the adverse effects caused by bad sample data are eliminated. The normalization formula is: ; In the above formula, is the normalized value; is the value to be normalized; , are the minimum and maximum values ​​respectively.

[0020] Then the optimal value is solved by the POA algorithm (Pelican Algorithm), and the output value obtained is the predicted air flow value; The specific method of the POA algorithm is as follows: The first step is to use formula (a) to randomly initialize the population individuals according to the upper and lower limits of the given problem variables (i.e., air compressor exhaust temperature, air compressor operating current, exhaust pressure, and air tank pressure); , , β=1,2,3,4…M, (a-1); In formula (a-1), For the The β-th dimension position of the pelican; To find the lower limit of the βth dimension of the problem; rand is a random number in the range [0, 1]; is the upper limit of the βth dimension of the problem to be solved; S is the population size of pelicans; M is the dimension of the problem to be solved.

[0021] The population of pelicans is shown in formula (a-2). Each row of the matrix represents the position of each pelican, and each column represents the value of the problem variable.

[0022] (a-2); Where X is the population matrix of pelicans; For the Only the Pelican's location.

[0023] In addition, in the POA algorithm, the objective function value of the pelican population can be expressed by the objective function value vector: (a-3); In the above formula, F is the objective function vector of the pelican population; For the The objective function value for the pelican.

[0024] Then it enters the position update, which includes the exploration phase and the development phase. The exploration phase follows the following formula: (a-4); In the formula, In the first stage The updated position of the pelican in the βth dimension; is the position of the prey in the βth dimension; I is a random number, which takes the value of 1 or 2; is the objective function value of the prey. Parameter I affects the global search capability of the POA algorithm. After the exploration is completed, the objective function value of the new position and the objective function value of the prey are determined to be larger or smaller. The smaller one is selected. The update method is as follows: ; (a-5) In the formula, For the The new location of the pelican, is the objective function value in the second stage.

[0025] The formula for the development phase is: ; (a-6) For the second stage The updated position of the pelican in the βth dimension; R is a constant, R=0.2; for The neighborhood radius is , e is the current iteration number; E is the maximum iteration number. At this stage, the effective update is also used to accept or reject the new pelican position. The process can be described by formula (a-7): ; (a-7) In the formula, For the The new location of the pelican, is the objective function value in the second stage.

[0026] After all individuals in the population have been updated in the above two stages, the optimal solution of the objective function is obtained, and the iteration is repeated until it is completely completed, and finally the global optimal solution to the problem is obtained.

[0027] In addition, in this embodiment, the air flow value X and the parameter data Y related to the air flow n The correlation coefficient calculation formula is: ; in The value range of is [0, 1.0], the range of strong correlation is [0.9-1.0], the range of relatively strong correlation is [0.6-0.9), the range of medium correlation is [0.4-0.6), and the range of no correlation is [0 -0.4); according to calculations, the data with strong correlation in this embodiment are the air compressor exhaust temperature, the air compressor operating current, the exhaust pressure and the air tank pressure.

[0028] After obtaining the predicted air flow value, this embodiment calculates the predicted power of each air compressor according to the formula: P = k×Q×s / 60; wherein P represents the output power of the air compressor in kilowatts; k is a constant whose value is 1.471; Q represents the air flow rate in cubic meters per minute; and s is the rotation speed of the air compressor in revolutions per minute.

[0029] S300, calculating a similarity index based on the predicted power and load power of each air compressor, and determining an air compressor that can meet the load usage requirement when the similarity index is a positive value.

[0030] The similarity index between the predicted power and load power calculation of each air compressor is calculated as follows: Determine the operation period T, and the predicted power of each air compressor in the operation period is recorded as P N {P N1 , P N2 …, P Ni}, where P Ni is the predicted power of the air compressor in period i, and the load power in one operating cycle is P L {P L1 , P L2 …, P Li}, then the similarity index between the two is R(P N , P L ) is calculated as follows: ; ; ; ; In the above formula, is the covariance between the predicted power and load power of each air compressor; and is the variance of the predicted power and load power of each air compressor, It is the average value of the predicted power of the air compressor in one operating cycle; is the average value of load power in one operating cycle; since R(P N , P L) has a value range of [-1,1], and its value can accurately describe the tracking matching degree between the load and the air compressor: when the load curve and the air compressor curve have the same changing trend, R(P N , P L ) is a positive value; when the load curve changes in the opposite direction to the air compressor curve, R(P N , P L ) is a negative value; when the load curve and the air compressor curve have a trend close to a completely monotonic correlation, | R(P N , P L )| will increase. When the correlation is completely monotonically related, | R(P N , P L )|=1; when there is no correlation between the load curve and the air compressor curve, R(P N , P L ))=0, therefore, in this embodiment, in the same period of time, the air compressor when the load curve and the air compressor curve have the same changing trend is selected as the load, that is, this part of the air compressors can meet the requirements of load use.

[0031] After determining the air compressor that can be used, select the corresponding air compressor to work according to the operating status of the air compressor and the load. The specific methods are as follows: According to the similarity index, the number of air compressors that can meet the load usage requirements is determined as Z, and the mathematical model of air compressor system energy consumption is established, and its objective function is: ; In the above formula, the operating energy consumption of the EC air compressor unit; is a binary variable, representing the operating status of the jth air compressor in the time period; is the predicted air flow value of the j-th air compressor in period t; The operating energy consumption of the j-th air compressor in period t; ω is the operating energy consumption of the j-th air compressor from shutdown to startup; is the air compressor running time; Z is the air compressor that can meet the load usage requirements.

[0032] in The relationship with the predicted air flow value is as follows: ; a j 、b j 、c j ——Fitting parameters of the energy consumption characteristic curve of the jth air compressor.

[0033] Constraints are established during scheduling, including load balance constraints, gas production constraints, air compressor start and stop times constraints, and binary variable constraints.

[0034] The load balancing constraints are: ; In the above formula, is the gas consumption required in period t.

[0035] The air output of the air compressor varies under different operating conditions. When the speed is constant and the air output is lower than a certain value, the air compressor will surge. Therefore, the minimum air output of the air compressor is the surge flow rate; the maximum flow rate is generally the rated flow rate of the air compressor, which is expressed as follows: The gas production constraint is ; In the above formula, is the minimum gas output of the jth air compressor, i.e., surge flow rate; is the minimum gas output of the jth air compressor, that is, the rated flow rate.

[0036] The number of starts and stops of the air compressor is limited, and the expression is as follows: ;M j is the maximum allowed number of shutdowns per day for the j-th air compressor.

[0037] Binary variable constraints: Indicates that the air compressor is in operation. Represents that the air compressor is in shutdown state, and its expression is as follows: .

[0038] Embodiment 2: The present invention also provides a computer device, which includes a memory and a processor. The memory is used to store computer instructions, and the processor executes the optimization scheduling method of the air compressor system by executing the computer instructions.

[0039] Embodiment 3: The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the optimization scheduling method for an air compressor system.

[0040] The present invention adopts an improved prediction algorithm, selects the air compressor that best matches the load demand by calculating the similarity index, and makes an accurate prediction based on the operating mechanism and historical data of the air compressor, thereby improving the accuracy and reliability of the prediction algorithm.

[0041] Optimize resource allocation: When selecting an air compressor, the present invention not only considers the parameters of the air compressor, but also comprehensively considers the operating status and load requirements of the air compressor, thereby realizing reasonable selection and scheduling of the air compressor, avoiding the situation of excess or insufficient capacity of the air compressor, optimizing resource allocation, and improving production efficiency.

[0042] Reducing energy consumption: Under the premise of ensuring load usage requirements, the present invention minimizes the operating energy consumption of the air compressor unit, and achieves efficient use of energy and reduces energy consumption by optimizing the selection and scheduling of the air compressor.

[0043] Improve system energy efficiency: The air compressor system energy consumption mathematical model of the present invention takes into account the energy consumption characteristic curve and operating status of the air compressor, and improves the energy efficiency of the air compressor system by optimizing the start and stop and operating status of the air compressor.

[0044] High flexibility: The scheduling method of the present invention can flexibly adjust the selection and scheduling strategy of the air compressor according to different load requirements and environmental conditions, thereby improving the adaptability and flexibility of the system. In summary, compared with the prior art, the present invention has higher prediction accuracy, better resource allocation, lower energy consumption, higher system energy efficiency and higher flexibility. It is an advanced air compressor control method with important practical value.

[0045] For example, certain words are used in the specification and claims to refer to specific components. Those skilled in the art should understand that hardware manufacturers may use different terms to refer to the same component. This specification and claims do not use differences in names as a way to distinguish components, but use differences in the functions of components as the criteria for distinction. For example, "including" mentioned throughout the specification and claims is an open term, so it should be interpreted as "including but not limited to". "Approximately" means that within an acceptable error range, those skilled in the art can solve technical problems within a certain error range and basically achieve technical effects.

[0046] It should be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a product or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such product or system. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the product or system including the elements.

[0047] The above description shows and describes several preferred embodiments of the present invention, but as before, it should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the invention concept of this article through the above teachings or the technology or knowledge of the relevant field. And the changes and modifications made by those skilled in the art do not depart from the spirit and scope of the present invention, and should be within the scope of protection of the claims attached to the present invention.

Claims

1. An optimization scheduling method for an air compressor system, characterized in that: The steps include: S100, obtain the air flow value of each air compressor in a certain operation cycle, and the data is X{x1, x2, x3…x n }, and obtain the parameter data Y related to the air flow n Where n=1, 2, 3…, the specific data of the parameter data is Y n {Y n1 , Y n2 , Y n3 …Y nn }, calculate the data X{x1, x2, x3…x n } and Y n {Y n1 , Y n2 , Y n3 …Y nn }, select parameter data with strong correlation; S200, the parameter data with strong correlation is predicted by the POA-LSTM prediction model to obtain the predicted air flow value of each air compressor, and the predicted power of each air compressor is calculated according to the predicted air flow value; S300, calculating a similarity index based on the predicted power and load power of each air compressor, and determining an air compressor that can meet the load usage requirement when the similarity index is a positive value.

2. The optimization scheduling method of an air compressor system according to claim 1, characterized in that: Parameter data Y related to air flow in step S100 n Including compressed air leakage, exhaust pressure, air compressor operating current, air tank pressure, and air compressor exhaust temperature.

3. The optimization scheduling method of an air compressor system according to claim 1, characterized in that: Air flow value X and air flow related parameter data Y n The correlation coefficient calculation formula is: ； in The value range of is [0, 1.0], the range of strong correlation is [0.9-1.0], the range of moderate correlation is [0.6-0.9), the range of medium correlation is [0.4-0.6), and the range of no correlation is [0 -0.4).

4. The optimization scheduling method of an air compressor system according to claim 1, characterized in that: After obtaining the predicted air flow value, the predicted power of each air compressor is calculated according to the formula: P = k×Q×s / 60; where P represents the output power of the air compressor in kilowatts; k is a constant with a value of 1.471; Q represents the air flow in cubic meters per minute; and s is the speed of the air compressor in revolutions per minute.

5. The method for optimizing and scheduling an air compressor system according to claim 1, characterized in that: The similarity index between the predicted power and load power calculation of each air compressor is calculated as follows: Determine the operation period T, and the predicted power of each air compressor in the operation period is recorded as P N {P N1 , P N2 …, P Ni }, where P Ni is the predicted power of the air compressor in period i, and the load power in one operating cycle is P L {P L1 , P L2 …, P Li }, then the similarity index between the two is R(P N , P L ) is calculated as follows: ； ； ； ； In the above formula, is the covariance between the predicted power and load power of each air compressor; and is the variance of the predicted power and load power of each air compressor, It is the average value of the predicted power of the air compressor in one operating cycle; It is the average value of load power in one operating cycle.

6. The method for optimizing and scheduling an air compressor system according to claim 5, characterized in that: According to the similarity index, the number of air compressors that can meet the load usage requirements is determined as Z, and the mathematical model of air compressor system energy consumption is established, and its objective function is: ； In the above formula, the operating energy consumption of the EC air compressor unit; is a binary variable, representing the operating status of the jth air compressor in the time period; is the predicted air flow value of the jth air compressor in period t; The operating energy consumption of the j-th air compressor in period t; ω is the operating energy consumption of the j-th air compressor from shutdown to startup; is the air compressor running time; Z is the air compressor that can meet the load usage requirements.

7. The method for optimizing and scheduling an air compressor system according to claim 6, characterized in that: in The relationship between the predicted air flow value of the j-th air compressor in period t is as follows: ； a j , b j 、c j ——Fitting parameters of the energy consumption characteristic curve of the jth air compressor.

8. The method for optimizing and scheduling an air compressor system according to claim 6, characterized in that: Constraints are established during scheduling, including load balance constraints, gas production constraints, pipeline pressure constraints, air compressor start and stop times constraints, and binary variable constraints.

9. A computer device, characterized in that: The device includes a memory and a processor, the memory is used to store computer instructions, and the processor executes the computer instructions to execute the optimization scheduling method of an air compressor system described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the optimization scheduling method for an air compressor system as described in any one of claims 1-8.

Images