Compressed air flow distribution method and device of air compression station and electronic equipment

Abstract

The invention provides a compressed air flow distribution method and device of an air compression station and electronic equipment, and belongs to the technical field of industrial automation. The method comprises the steps of obtaining operation data of air compression station composition equipment, terminal equipment air source demand quantity and at least one control target, adopting a multi-target optimization algorithm, and analyzing the operation data by using a deep learning network according to the control target, and predicting the optimal operation parameters and predicted operation data of the air compression station composition equipment when the air source demand of the terminal equipment is met by adopting a weighted sum method for the control target, controlling the operation of the air compression station composition equipment according to the updated optimal operation parameters, and dynamically distributing compressed air flow to the terminal. By means of the method, the automation degree is improved, dynamic distribution and optimization of the compressed air flow of the air compression station can be achieved, the air source distribution efficiency and unit efficient dispatching are improved, and production interruption caused by improper flow distribution is effectively avoided.

Description

Technical Field

[0001] The present invention belongs to the technical field of industrial automation and relates to a compressed air flow distribution method, device and electronic equipment for an air compressor station. Background Art

[0002] With the continuous development of industrial automation and intelligent equipment, traditional air compressor stations face many challenges in flow distribution, monitoring, and management. In particular, existing flow distribution methods are typically based on fixed rules, making it difficult to dynamically adjust according to real-time demand, resulting in energy waste or insufficient air supply. Furthermore, the control method relies on the air compressor's own system for single-unit control, which is not conducive to the coordinated scheduling of multiple units. Furthermore, when operating conditions change, the existing system is unable to quickly adjust flow distribution, resulting in low operating efficiency, insufficient support for the overall energy efficiency optimization of the air compressor station, and difficulty in effectively reducing energy consumption. In short, existing technologies make it difficult to achieve efficient and accurate flow control and optimization in air compressor stations. Summary of the Invention

[0003] The purpose of the present invention is to address the above-mentioned problems in the existing technology and to propose a compressed air flow distribution method, device and electronic equipment for an air compression station.

[0004] The object of the present invention can be achieved by the following technical solutions: A method for distributing compressed air flow in an air compression station, comprising:

[0005] Obtaining the operating data of the components of the air compression station, the air source demand of the terminal equipment and at least one control target;

[0006] Adopting a multi-objective optimization algorithm, using a deep learning network to analyze the operating data according to the control objective, and using a weighted sum method to predict the optimal operating parameters and predicted operating data of the components of the air compression station when meeting the gas source demand of the terminal equipment;

[0007] Controlling the operation of the components of the air compressor station according to the optimal operating parameters to obtain real-time operating data, analyzing the real-time operating data according to the predicted operating data, and performing rewards or penalties, wherein the deep learning network dynamically updates weights according to the rewards or penalties;

[0008] The optimal operating parameters of the components of the air compression station are updated according to the updated weight prediction, the operation of the components of the air compression station is controlled according to the updated optimal operating parameters, and the compressed air flow is dynamically allocated to the terminal.

[0009] As an optional embodiment of the present invention, a multi-objective optimization algorithm is used to analyze the operating data using a deep learning network according to the control target, and a weighted sum method is used to predict the optimal operating parameters of the components of the air compressor station when the gas source demand of the terminal device is met. The method includes:

[0010] The control objective includes minimizing the total energy consumption of the air compressor station while meeting the gas source demand of the terminal equipment;

[0011] Obtain the relationship between the power of each compressor and the corresponding output compressed air flow rate;

[0012] Based on the control target, the optimal operating time and optimal air supply volume of each compressor are weightedly predicted according to the relationship and the relationship of minimum total energy consumption of the air compression station while meeting the air source demand of the terminal equipment and minimizing the total energy consumption of the air compression station.

[0013] As an optional implementation scheme of the present invention, the control target also includes: the optimal gas supply volume is greater than or equal to the gas source demand of the terminal equipment.

[0014] As an optional embodiment of the present invention, a multi-objective optimization algorithm is used to analyze the operating data using a deep learning network according to the control target, and a weighted sum method is used to predict the optimal operating parameters of the components of the air compressor station when the air source demand of the terminal device is met. The method also includes:

[0015] The control target also includes minimizing the pressure fluctuation deviation of the air storage tank when the pressure of the air storage tank in the air compression station meets the compressed air supply process;

[0016] Construct a pressure fluctuation dynamic model based on the gas tank volume, inlet flow rate and outlet flow rate;

[0017] Based on the control target, when the internal pressure of the gas storage tank fluctuates, the minimum pressure fluctuation deviation is predicted by weighted prediction according to the pressure fluctuation dynamic model when the gas source demand of the terminal device is met and the pressure of the gas storage tank is within a preset pressure range.

[0018] As an optional embodiment of the present invention, the deep learning network includes:

[0019] Inputting the control target into the state space of the deep learning network;

[0020] Inputting the weight corresponding to the control target into the dynamic space of the deep learning network;

[0021] A corresponding reward function is constructed according to the state space and the dynamic space, and a deep learning network for the optimal operating parameters of the air compressor station components is constructed according to the state space, the dynamic space, and the reward function.

[0022] As an optional embodiment of the present invention, it also includes:

[0023] Acquire all historical optimal operating parameters and corresponding historical operating data within a preset time range at a fixed frequency, perform global strategy correction on the historical optimal operating parameters and the historical operating data, and obtain an optimal operating parameter set and a corresponding operating data set;

[0024] The optimal operating parameter set and the operating data set are input into the deep learning network and trained. When new operating data is received, the trained deep learning network is used to predict the optimal operating parameters and predicted operating data of the air compression station components when meeting the gas source demand of the terminal equipment.

[0025] As an optional embodiment of the present invention, performing a global strategy correction on the historical optimal operating parameters and the historical operating data to obtain an optimal operating parameter set and a corresponding operating data set includes:

[0026] Taking the historical optimal operating parameters and the corresponding historical operating data as a set of data, and analyzing whether each set of data meets the control target;

[0027] The fitness of multiple sets of data that meet the control target is sorted from high to low, a preset number of sets of data are obtained from the sorting results, and integrated into an optimal operating parameter set and a corresponding operating data set.

[0028] As an optional embodiment of the present invention, real-time operation data is analyzed based on the predicted operation data to perform rewards or penalties, and the deep learning network dynamically updates weights based on the rewards or penalties; comprising:

[0029] The predicted operating data and real-time operating data of the control target are analyzed. If the analysis result meets the preset rules, corresponding rewards or penalties are given, and the weight of the control target is updated according to the rewards or penalties.

[0030] The present invention also provides a compressed air flow distribution device for an air compression station, comprising:

[0031] A data acquisition module is used to obtain the operating data of the components of the air compression station, the gas source demand of the terminal equipment and at least one control target;

[0032] a data analysis module for adopting a multi-objective optimization algorithm, using a deep learning network to analyze the operating data according to the control target, and using a weighted sum method to predict the optimal operating parameters and predicted operating data of the components of the air compression station when meeting the gas source demand of the terminal device;

[0033] a feedback module for controlling the operation of the components of the air compressor station according to the optimal operating parameters, obtaining real-time operating data, analyzing the real-time operating data according to the predicted operating data, and performing rewards or penalties, wherein the deep learning network dynamically updates weights according to the rewards or penalties;

[0034] The flow distribution module is used to update the optimal operating parameters of the air compression station components according to the updated weight prediction, control the operation of the air compression station components according to the updated optimal operating parameters, and dynamically distribute the compressed air flow to the terminal.

[0035] The present invention further provides an electronic device, comprising:

[0036] processor;

[0037] a memory for storing processor-executable instructions;

[0038] Wherein, the processor is configured to implement the above-mentioned compressed air flow distribution method of the air compression station when executing the executable instructions.

[0039] Compared with the existing technology, the present invention adopts a multi-objective optimization algorithm combined with a deep learning network to predict the optimal operating parameters, thereby improving the degree of automation, being able to realize the dynamic allocation and optimization of the compressed air flow of the air compressor station, improving the gas source distribution efficiency and efficient scheduling of the unit, and improving the response speed and stability of the air compressor station system, effectively avoiding production interruptions caused by improper flow distribution. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a flow chart of a compressed air flow distribution method for an air compression station according to an embodiment of the present invention;

[0041] Figure 2 This is a block diagram of a compressed air flow distribution device for an air compression station according to an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The following are specific embodiments of the present invention and the accompanying drawings to further describe the technical solutions of the present invention, but the present invention is not limited to these embodiments.

[0043] Example 1

[0044] An air compression station, or air compression station, is a system used to provide compressed air in industrial settings. Its core components include air compressors, air storage tanks, post-processing equipment, piping systems, and corresponding control and monitoring equipment. Compressed air is a vital form of energy in many industrial fields and is widely used to drive equipment and control pneumatic systems.

[0045] Flow distribution is the process of allocating compressed air to different ports or production lines based on demand. In an air compression station, flow distribution must meet the real-time needs of different equipment while optimizing the energy efficiency of the entire system.

[0046] With the continuous development of industrial automation and intelligent equipment, this embodiment proposes a compressed air flow distribution method for an air compressor station, aiming to solve the technical problems reflected in the background technology, such as Figure 1 As shown, a compressed air flow distribution method for an air compression station includes:

[0047] S1, obtaining the operating data of the components of the air compression station, the air source demand of the terminal equipment and at least one control target;

[0048] S2, using a multi-objective optimization algorithm, using a deep learning network to analyze the operating data according to the control target, and using a weighted sum method to predict the optimal operating parameters and predicted operating data of the components of the air compressor station when meeting the gas source demand of the terminal device;

[0049] S3, controlling the operation of the components of the air compressor station according to the optimal operating parameters, obtaining real-time operating data, analyzing the real-time operating data according to the predicted operating data, and performing rewards or penalties, wherein the deep learning network dynamically updates weights according to the rewards or penalties;

[0050] S4, updating the optimal operating parameters of the components of the air compression station according to the updated weight prediction, controlling the operation of the components of the air compression station according to the updated optimal operating parameters, and dynamically allocating the compressed air flow to the terminal.

[0051] Sensors installed on various components of the air compression station, such as flow sensors, pressure sensors, and temperature sensors, collect operating parameters and corresponding data in real time. This data is then preprocessed and feature extracted through time alignment, noise filtering, and outlier correction. Real-time data also includes the air supply demand of terminal equipment. Control objectives are set by the operator. For example, minimizing system energy consumption and ensuring a stable compressed air reserve are key objectives. This means that while meeting the air supply demand of terminal equipment, these control objectives must also be considered comprehensively.

[0052] The number of control targets is one or more. In this embodiment, a deep learning agent is used to analyze the operating data by comprehensively considering the control targets through a built-in multi-objective optimization algorithm, initialize the weights of each control target, and use the weighted sum method to predict the optimal action of the component equipment, that is, the optimal operating parameters of the component equipment. And the predicted operating data of the component equipment when working according to the optimal operating parameters. The operating parameters of the component equipment are adjusted to the predicted optimal operating parameters, and the corresponding real-time operating data are recorded. In this embodiment, a dynamic weight adjustment strategy based on DQN is also designed. According to the aforementioned predicted operating data, the real-time operating data is analyzed to evaluate the performance of the component equipment of the air compressor station, such as energy efficiency utilization, gas source demand satisfaction, gas tank pressure stability, etc. A reward or penalty is given according to the performance through a reward function to guide the adjustment of the weights of each control target of the component equipment in the next step.

[0053] After the weight adjustment, the optimal operating parameters of the component equipment are updated accordingly. The component equipment will operate according to the updated optimal operating parameters and dynamically distribute compressed air flow to the terminal. By using a deep learning agent to analyze the operating data of the current air compressor station component equipment, the optimal operating parameters and corresponding predicted operating data of the air compressor station component equipment are predicted based on the control target analysis. For each component equipment operation data, it will be analyzed based on the latest predicted operation data, and rewards or penalties will be given. According to the rewards or penalties, the weights corresponding to each control target are adjusted to obtain the corresponding optimal operating parameters, avoiding the limitations of fixed weights in the traditional weighted sum method. The operating data obtained according to the optimal operating parameters meets the air source demand of the terminal equipment while meeting the control target. This cycle repeats over time. The deep learning agent uses its own convergence characteristics to gradually adapt to the optimization of compressed air flow distribution under various working conditions.

[0054] Preferably, a multi-objective optimization algorithm is adopted, and a deep learning network is used to analyze the operating data according to the control target, and a weighted sum method is used to predict the optimal operating parameters of the components of the air compressor station when the gas source demand of the terminal equipment is met, including: the control target includes minimizing the total energy consumption of the air compressor station when the gas source demand of the terminal equipment is met; obtaining the relationship between the power of each compressor and the corresponding output compressed air flow rate; based on the control target, according to the relationship and the relationship between the minimum total energy consumption of the air compressor station, weighted prediction is made of the optimal operating time and optimal air supply of each compressor when the gas source demand of the terminal equipment is met and the total energy consumption of the air compressor station is minimized.

[0055] In this embodiment, one of the control objectives is to minimize the energy consumption of the air compression station, that is, to minimize the energy consumption of the air compression station while meeting the gas source demand of all terminal devices. The following formula is used:

[0056]

[0057] Among them, E Total is the total energy consumption of the air compression station; P i is the power of the i-th air compressor; P i With the output flow Q i It is a nonlinear relationship, which can be fitted through the air compressor characteristic curve as follows:

[0058]

[0059] Where: a, b, c are constants, and the i-th air compressor passes through multiple groups of power P i With the output flow Q i The experimental results show that the air compressor characteristic curve fitting process is as follows: Get the power P of the air compressor during the operation of the air compressor i and output flow Q i The relationship diagram or scatter plot of power P i and output flow Q i Fitted into a quadratic function. Output flow Q i That is, the actual gas supply volume of the i-th terminal device, in m 3 / min; T i is the operating time of the i-th air compressor. For this control target, the weighted sum of the formula can be obtained by optimizing the subsequent start and stop state T of each air compressor. i and flow Q i , reducing the total energy consumption of the air compressor station. i That is the optimal running time, Q i That is, the optimal air supply, which means that when the air compressor operating time is the optimal operating time and the actual air supply is the optimal air supply, the energy consumption of the air compression station is the minimum.

[0060] Preferably, the control target further includes: the optimal gas supply volume is greater than or equal to the gas source demand of the terminal equipment.

[0061] Each terminal device in the air compressor station, such as production lines and pneumatic tools, has a specific air source requirement. The control goal is to ensure that the requirements of each terminal are met. This constraint can be expressed as the following inequality:

[0062]

[0063] D i is the gas source demand of the i-th terminal device, and m is the total number of terminal devices. Considering the relationship between the gas supply volume and the gas source demand of the terminal device, when the gas supply volume is equal to the gas source demand of the terminal device, it just meets the terminal demand and does not cause waste of compressed air flow. However, due to the need to consider the system energy consumption of the air compressor station and the pressure inside the gas tank, Qi It is not necessarily equal to the gas source demand of the terminal equipment to minimize the system energy consumption. From the above, we can know that the system energy consumption is also related to the running time T i Therefore, the control goal is to ensure that when the air compressor station distributes the flow, the optimized Q corresponding to each terminal device i It can meet the gas source demand of each terminal device, and will not cause production interruption due to insufficient supply. In addition, the gas tank maintains a certain pressure range during the process of distributing compressed air flow.

[0064] Preferably, a multi-objective optimization algorithm is used, a deep learning network is used to analyze the operating data according to the control target, and a weighted sum method is used to predict the optimal operating parameters of the components of the air compression station when the gas source demand of the terminal device is met. The method also includes:

[0065] The control target also includes minimizing the pressure fluctuation deviation of the gas tank when the pressure of the gas tank in the air compression station meets the compressed air supply process; constructing a pressure fluctuation dynamic model based on the volume, inlet flow rate and outlet flow rate of the gas tank; based on the control target, when the pressure inside the gas tank fluctuates, weighted prediction is made according to the pressure fluctuation dynamic model to determine the minimum pressure fluctuation deviation when the gas source demand of the terminal device is met and the pressure of the gas tank is within a preset pressure range.

[0066] The air storage tanks in an air compression station need to maintain a certain pressure range during the compressed air supply process. Too low a pressure will result in insufficient air supply, while too high a pressure will waste energy and increase equipment load. This can be expressed by the following constraints:

[0067] P min ≤P 储罐 ≤P max

[0068] Pressure fluctuations are mainly caused by sudden load changes and compressor response lags. Its dynamic model can be simplified as follows:

[0069]

[0070] Where V is the volume of the tank, unit: m 3 ;Q in , Q out They are respectively the inlet flow rate and the outlet flow rate, unit: m 3 / min; the control objective is to optimize and minimize the pressure deviation:

[0071] min P 储罐 -Pave

[0072] Among them, P aveis the average pressure of the system, which can be obtained by conventional technical means in this field and will not be described in detail here. For the above control objectives, the above control objectives are comprehensively considered when allocating the compressed air flow, and the multi-objective optimization is converted into a single-objective optimization, and the weighted sum method is used to process it:

[0073]

[0074] Where ω1, ω2, ω3 are the weights of each control target, E Total is the total energy consumption of the air compression station, Indicates the satisfaction degree of gas source demand; |P 储罐 -P ave | represents the penalty term for pressure fluctuations in the gas tank. The result F obtained is the optimal operating parameters and predicted operating data of the components of the air compressor station when the gas source demand of the terminal device is met. In the scenario of compressed air flow distribution in the air compressor station, based on the operating parameters of each compressor, the operating data of the user end and the intermediate transmission and processing link, all control objectives are comprehensively considered to dynamically search for the optimal operating parameters and predicted operating data. It should be noted that, in addition to the control objectives in this embodiment, the operator can also add or change the control objectives during actual application. However, no matter how many control objectives there are, they can be converted into single-objective optimization through a multi-objective optimization algorithm, and the weighted sum method is used to obtain the optimal operating parameters and predicted operating data of the corresponding components that meet the gas source demand of the terminal device.

[0075] Preferably, the deep learning network includes:

[0076] The control target is input into the state space of the deep learning network; the weight corresponding to the control target is input into the dynamic space of the deep learning network; a corresponding reward function is constructed according to the state space and the dynamic space, and a deep learning network for the optimal operating parameters of the air compressor station components is constructed according to the state space, the dynamic space, and the reward function.

[0077] In this embodiment, the deep learning network is mainly based on the deep Q network, and a dynamic weight adjustment strategy based on DQN is designed. Specifically, the control target is input into the state space S t In the middle, the gas tank pressure P 储罐 , Power of each air compressor P i , terminal equipment gas source demand and total energy consumption E Total Action space A t Where ω1, ω2, ω3 are weight vectors, which are used to adjust the weight vector within a certain range. The reward function reflects the optimization effect of multiple objectives and is defined as:

[0078]

[0079] Where: λ is the stability bonus coefficient, which is used to amplify the positive incentives brought by long-term stability (e.g., λ = 1); Stability_Bonus is an additional bonus that can be dynamically calculated based on the pressure fluctuation of the tank. For example, when the standard deviation of the pressure fluctuation is below the threshold for 10 consecutive minutes, an additional bonus of +10 is awarded. Furthermore, the deep learning network used to predict the most reliable operating parameters of the component equipment and the predicted operating data is:

[0080]

[0081] S t is the current state, A t is the current sampling action, F t is the reward value, α is the learning rate, and γ is the discount factor, which measures the impact of future rewards.

[0082] The deep Q network helps the air compressor station system to make decision optimization in a complex, multi-objective constrained environment, which can improve the operating efficiency of the air compressor station, reduce energy consumption, and ensure the stability of the system.

[0083] Preferably, it also includes:

[0084] All historical optimal operating parameters and corresponding historical operating data within a preset time range are obtained at a fixed frequency, and global strategy corrections are performed on the historical optimal operating parameters and the historical operating data to obtain an optimal operating parameter set and a corresponding operating data set; the optimal operating parameter set and the operating data set are input into the deep learning network and trained, and when new operating data is received, the trained deep learning network is used to predict the optimal operating parameters and predicted operating data of the air compressor station components when the gas source demand of the terminal equipment is met.

[0085] In this embodiment, a global strategy correction is performed on the historical optimal operating parameters within a certain time period at a fixed frequency of once a week to obtain the optimal operating parameter set and the corresponding operating data set for a certain time period, and use them to train the deep learning network. The deep learning network uses its own self-learning characteristics to learn the characteristics of the data in the optimal operating parameter set and the corresponding operating data set, and continuously iterates the strategy. When the trained deep learning network receives real-time operating data, it will become more and more accurate in predicting the optimal operating parameters and predicted operating data of the equipment that makes up the air compressor station, that is, the deep learning network performs regular optimization. It should be noted that the preset time range is set by the user according to the learning situation of the deep learning agent, and is not limited here.

[0086] Preferably, performing global strategy correction on the historical optimal operating parameters and the historical operating data to obtain an optimal operating parameter set and a corresponding operating data set includes:

[0087] The historical optimal operating parameters and the corresponding historical operating data are taken as a set of data, and each set of data is analyzed to see whether it meets the control target; the fitness of multiple sets of data that meet the control target is sorted from high to low, and a preset number of sets of data are obtained from the sorting results, and integrated into an optimal operating parameter set and a corresponding operating data set.

[0088] When predicting the optimal operating parameters, there will be corresponding predicted operating data. After that, based on whether the real-time operating data meets the control target, rewards or penalties will be given based on the analysis of the real-time operating data according to the predicted operating data. The overall strategy revision is to improve the performance of the air compressor station. Therefore, when obtaining the historical optimal operating parameters and the corresponding historical operating data, they can be used as a group of data, and each group of data can be analyzed one by one to see whether it meets the control target. If not, it will be ignored. If it is satisfied, it will be retained. For the multiple groups of data retained, the results F corresponding to each group of data are sorted by fitness. That is, among the multiple groups of historical optimal operating parameters and corresponding historical operating data that meet the control target, a preset number of groups of historical optimal operating parameters and corresponding historical operating data are selected according to the fitness from high to low, and integrated into the optimal operating parameter set and the corresponding operating data set.

[0089] Preferably, the real-time operation data is analyzed according to the predicted operation data, rewards or penalties are performed, and the deep learning network dynamically updates weights according to the rewards or penalties; comprising:

[0090] The predicted operating data and real-time operating data of the control target are analyzed. If the analysis result meets the preset rules, corresponding rewards or penalties are given, and the weight of the control target is updated according to the rewards or penalties.

[0091] The above example shows that additional rewards can be dynamically calculated based on the pressure fluctuation of the tank. For example, when the pressure fluctuation standard deviation is below the threshold for 10 consecutive minutes, an additional reward of +10 is given. In actual application, users can also set other rules to compare and analyze the predicted operation data with the real-time operation data. For example, in the real-time operation data, the total energy consumption E total Compared with the total energy consumption E in the predicted operation data total It is lower, indicating that the total energy consumption of the system is actually lower than the predicted one. total Each time a component device performs an action based on the predicted optimal operating parameters, the deep learning network will reward or penalize it based on the analysis results, and further adjust the next optimal operating parameters based on the reward or penalty. Over time, the strategy of the intelligent multi-objective optimization algorithm will gradually converge and gradually adapt to the optimal compressed air flow distribution under various working conditions.

[0092] By analyzing real-time operating data against predicted data, the health status of corresponding equipment can be determined and potential failures can be predicted. Remote control is also possible, allowing managers to remotely start and stop equipment, modify parameters, and schedule tasks through the platform, reducing manual intervention and control in the air compressor station system.

[0093] For the computing functions involved, edge computing can be used to send part of the computing functions of the deep learning network to the device side, reducing the burden on the central server in the air compression station.

[0094] Through the above method, a multi-objective optimization algorithm is combined with a deep learning network to predict the optimal operating parameters, which improves the degree of automation, can realize the dynamic allocation and optimization of the compressed air flow of the air compressor station, improve the gas source distribution efficiency and efficient scheduling of the unit, and improve the response speed and stability of the air compressor station system, effectively avoiding production interruptions caused by improper flow distribution.

[0095] Example 2

[0096] Based on the same principle as the above method, a compressed air flow distribution device 100 for an air compression station is also proposed. Figure 2 Shown, including:

[0097] The data acquisition module 110 is used to obtain the operating data of the components of the air compression station, the gas source demand of the terminal equipment and at least one control target;

[0098] The data analysis module 120 is configured to employ a multi-objective optimization algorithm to analyze the operating data using a deep learning network according to the control objectives, and to use a weighted sum method to predict the optimal operating parameters and predicted operating data of the components of the air compressor station when the air source demand of the terminal device is met.

[0099] A feedback module 130 is configured to control the operation of the components of the air compressor station according to the optimal operating parameters, obtain real-time operating data, analyze the real-time operating data according to the predicted operating data, and perform rewards or penalties, wherein the deep learning network dynamically updates weights according to the rewards or penalties;

[0100] The flow distribution module 140 is used to update the optimal operating parameters of the air compression station components according to the updated weight prediction, control the operation of the air compression station components according to the updated optimal operating parameters, and dynamically distribute the compressed air flow to the terminal.

[0101] Example 3

[0102] Furthermore, an electronic device is proposed, comprising:

[0103] processor;

[0104] a memory for storing processor-executable instructions;

[0105] Wherein, the processor is configured to implement the compressed air flow distribution method of the air compression station described in Example 1 when executing the executable instructions.

[0106] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the various components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0107] In addition, it should be noted that the descriptions of "first", "second", "one", etc. in the present invention are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly defined. The terms "connected", "fixed", etc. should be understood in a broad sense. For example, "fixed" can be a fixed connection, a detachable connection, or an integral whole; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances.

[0108] In addition, the technical solutions between the various embodiments of the present invention can be combined with each other, but it must be based on the fact that ordinary technicians in this field can implement it. When the combination of technical solutions is mutually contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0109] The specific embodiments described herein are merely illustrative of the spirit of the present invention. Persons skilled in the art may make various modifications, additions, or substitutions to the described specific embodiments without departing from the spirit of the present invention or exceeding the scope of the appended claims.

Claims

1. A method for distributing compressed air flow in an air compression station, characterized in that: include: Obtaining the operating data of the components of the air compression station, the air source demand of the terminal equipment and at least one control target; Adopting a multi-objective optimization algorithm, using a deep learning network to analyze the operating data according to the control objective, and using a weighted sum method to predict the optimal operating parameters and predicted operating data of the components of the air compression station when meeting the gas source demand of the terminal equipment; Controlling the operation of the components of the air compressor station according to the optimal operating parameters to obtain real-time operating data, analyzing the real-time operating data according to the predicted operating data, and performing rewards or penalties, wherein the deep learning network dynamically updates weights according to the rewards or penalties; The optimal operating parameters of the components of the air compression station are updated according to the updated weight prediction, the operation of the components of the air compression station is controlled according to the updated optimal operating parameters, and the compressed air flow is dynamically allocated to the terminal.

2. The compressed air flow distribution method of an air compression station according to claim 1, characterized in that: A multi-objective optimization algorithm is used to analyze the operating data using a deep learning network according to the control target, and a weighted sum method is used to predict the optimal operating parameters of the components of the air compressor station when the air source demand of the terminal equipment is met. The parameters include: The control objective includes minimizing the total energy consumption of the air compressor station while meeting the gas source demand of the terminal equipment; Obtain the relationship between the power of each compressor and the corresponding output compressed air flow rate; Based on the control target, the optimal operating time and optimal air supply volume of each compressor are weightedly predicted according to the relationship and the relationship of minimum total energy consumption of the air compression station while meeting the air source demand of the terminal equipment and minimizing the total energy consumption of the air compression station.

3. The compressed air flow distribution method of an air compression station according to claim 2, characterized in that: The control target also includes: the optimal gas supply volume is greater than or equal to the gas source demand of the terminal equipment.

4. The compressed air flow distribution method of an air compression station according to claim 2, characterized in that: A multi-objective optimization algorithm is used to analyze the operating data using a deep learning network according to the control target, and a weighted sum method is used to predict the optimal operating parameters of the components of the air compressor station when the air source demand of the terminal device is met. The method also includes: The control target also includes minimizing the pressure fluctuation deviation of the air storage tank when the pressure of the air storage tank in the air compression station meets the compressed air supply process; Construct a pressure fluctuation dynamic model based on the gas tank volume, inlet flow rate and outlet flow rate; Based on the control target, when the internal pressure of the gas storage tank fluctuates, the minimum pressure fluctuation deviation is predicted by weighted prediction according to the pressure fluctuation dynamic model when the gas source demand of the terminal device is met and the pressure of the gas storage tank is within a preset pressure range.

5. The compressed air flow distribution method of an air compression station according to claim 1, characterized in that: The deep learning network includes: Inputting the control target into the state space of the deep learning network; Inputting the weight corresponding to the control target into the dynamic space of the deep learning network; A corresponding reward function is constructed according to the state space and the dynamic space, and a deep learning network for the optimal operating parameters of the air compressor station components is constructed according to the state space, the dynamic space, and the reward function.

6. The compressed air flow distribution method of an air compression station according to claim 5, characterized in that: Also includes: Acquire all historical optimal operating parameters and corresponding historical operating data within a preset time range at a fixed frequency, perform global strategy correction on the historical optimal operating parameters and the historical operating data, and obtain an optimal operating parameter set and a corresponding operating data set; The optimal operating parameter set and the operating data set are input into the deep learning network and trained. When new operating data is received, the trained deep learning network is used to predict the optimal operating parameters and predicted operating data of the air compression station components when meeting the gas source demand of the terminal equipment.

7. The compressed air flow distribution method of an air compression station according to claim 6, characterized in that: Performing a global strategy correction on the historical optimal operating parameters and the historical operating data to obtain an optimal operating parameter set and a corresponding operating data set includes: Taking the historical optimal operating parameters and the corresponding historical operating data as a set of data, and analyzing whether each set of data meets the control target; The fitness of multiple sets of data that meet the control target is sorted from high to low, a preset number of sets of data are obtained from the sorting results, and integrated into an optimal operating parameter set and a corresponding operating data set.

8. The compressed air flow distribution method of an air compression station according to claim 6, characterized in that: Analyzing real-time operation data according to the predicted operation data to perform rewards or penalties, and dynamically updating weights of the deep learning network according to the rewards or penalties; include: The predicted operating data and real-time operating data of the control target are analyzed. If the analysis result meets the preset rules, corresponding rewards or penalties are given, and the weight of the control target is updated according to the rewards or penalties.

9. A compressed air flow distribution device for an air compression station, characterized in that: include: A data acquisition module is used to obtain the operating data of the components of the air compression station, the gas source demand of the terminal equipment and at least one control target; a data analysis module for adopting a multi-objective optimization algorithm, using a deep learning network to analyze the operating data according to the control target, and using a weighted sum method to predict the optimal operating parameters and predicted operating data of the components of the air compression station when meeting the gas source demand of the terminal device; a feedback module for controlling the operation of the components of the air compressor station according to the optimal operating parameters, obtaining real-time operating data, analyzing the real-time operating data according to the predicted operating data, and performing rewards or penalties, wherein the deep learning network dynamically updates weights according to the rewards or penalties; The flow distribution module is used to update the optimal operating parameters of the air compression station components according to the updated weight prediction, control the operation of the air compression station components according to the updated optimal operating parameters, and dynamically distribute the compressed air flow to the terminal.

10. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; Wherein, the processor is configured to implement the compressed air flow distribution method of the air compression station as described in any one of claims 1 to 8 above when executing the executable instructions.

Images