Intelligent centralized control method and system for air compressor unit and storage medium

Abstract

The invention discloses an intelligent centralized control method and system for an air compressor unit and a storage medium, and the method comprises the steps: building a first correlation model which is used for representing the correlation between the air pressure fluctuation of each air supply node, and a second correlation model which is used for representing the correlation between the air flow temperature of a specific section in an air supply pipe network and the air pressure of each air supply node; meanwhile, corresponding fluctuation models and weight values are configured for all the air supply nodes, so that the system can predict the air pressure requirements of all the air supply nodes in the future set time period based on the current air pressure data and temperature data of all the air supply nodes, and then the air pressure requirements of all the air supply nodes in the future set time period are predicted by changing the conduction relation of the air supply nodes and the local temperature of the air supply pipe network. The air pressure of the air supply node is adjusted, the air demand at the air supply node is responded in advance, accurate control over the air pressure at the tail end of the air supply pipe network is achieved, and air pressure fluctuation at the tail end of the air supply pipe network is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of control of air compressor units in air compressor stations, and more specifically, to an intelligent centralized control method, system and storage medium for air compressor units. Background Art

[0002] In automated production lines such as automobile manufacturing, pneumatic systems can effectively improve production efficiency. Air compressors (air compressors) are the core equipment of the entire pneumatic system as air source power. During the operation of the pneumatic system, the air is first compressed by the air compressor, then dried, filtered and pressure-regulated, and finally supplied to various pneumatic terminals through the air supply network, such as cylinders, air motors, air claws, etc. In order to reduce the fluctuation range of air pressure at each pneumatic terminal and ensure production accuracy and efficiency, multiple air compressors are usually configured in an air compression station to achieve continuous operation and stable air supply. At the same time, at least one air storage tank is also configured in the air supply network, such as at the air compressor outlet, to buffer the air pressure changes in the air supply network.

[0003] In order to ensure the gas supply quality of the gas supply network and reduce the fluctuation amplitude of the air pressure in the network, a pressure sensor is usually installed in the gas tank. When it is detected that the pressure value in the gas tank is lower than the first set value, the air compressor is started. When the air pressure value in the gas tank is higher than the second set value, the air compressor is disconnected or the power of the air compressor is adjusted, so that the air pressure in the gas tank is finally maintained in the set range.

[0004] However, in actual applications, the inventors found that the existing control methods of air compressor units are not sufficient to meet the needs of high-precision pneumatic systems. For example, in ultra-precision air floating platforms, slight air pressure fluctuations at the end of the air supply network will cause the air floating platform to shake or even scratch the base. The main reason for the unstable air pressure at the pneumatic terminal and the tendency to slight fluctuations is that the air supply system responds too slowly to changes in the air pressure at the end.

[0005] Analysis has confirmed that the main reasons for the long response time of the pneumatic system include: (1) Transmission delay of control signal: When the pressure sensor in the gas storage tank detects the pressure change, the pressure change at the pneumatic terminal at the end of the gas supply network has already been very large. At this time, starting the air compressor through the controller to increase the pressure obviously cannot quickly solve the problem of insufficient air pressure at the end of the gas supply network; (2) Air flow delay in the gas supply network: Since air is highly compressible, the pressure change generated in the gas tank needs a certain delay before it can be transmitted to the end of the network through the pressure regulating valve. Especially when multiple pneumatic terminals at the end of the gas supply network are connected in parallel, if one of the pneumatic terminals is in a deflated state, the pneumatic output force of the remaining pneumatic terminals will be affected, and it will obviously take a certain amount of time to make up for this pressure loss.

[0006] In summary, how to shorten the air supply response time in the pneumatic system is the key to improving the air supply quality of the pneumatic system. Summary of the invention

[0007] In view of the problem that the air supply of the air compressor station is unstable in actual use, resulting in fluctuations in the air pressure at the end of the gas supply network, the first purpose of the present invention is to propose an intelligent centralized control method for air compressor units. Based on the air pressure fluctuation data at the end of the gas supply network, a gas supply model is generated in a targeted manner to control the air compressor units in the air compressor station to respond to the gas demand in advance. At the same time, through the linkage control of the gas supply pipelines between the pipelines and the adjustment of the gas state, the air pressure at the end of the gas supply network is precisely controlled, and the air pressure fluctuation at the end of the gas supply network is reduced. In order to realize the above-mentioned intelligent centralized control method for air compressor units, the second purpose of this application is to propose an intelligent centralized control system for air compressor units. Finally, it is proposed to protect a computer-readable storage medium loaded with a program module for realizing the intelligent centralized control method for air compressor units. The specific scheme is as follows: An intelligent centralized control method for an air compressor unit, based on the air compressor unit and the connected air supply network, comprises: Establishing and storing a first correlation model for characterizing the correlation relationship between the air pressure fluctuations of each air supply node; Establishing and storing a second correlation model for characterizing the correlation between the air flow temperature in a specific section of the air supply network and the air pressure at each air supply node; Obtain and generate a fluctuation model for predicting the pressure fluctuation trend of each gas supply node based on the historical air pressure data and temperature data of each gas supply node, and store each fluctuation model in association with the ID of each gas supply node; Based on the historical air pressure and temperature data of each gas supply node, the reliability of each fluctuation model is verified and a weight value is assigned to each fluctuation model; Obtain the air pressure data, temperature data of each air supply node at the current moment and the conduction status between each air supply node; According to the fluctuation model and weight value of each gas supply node, combined with the first correlation model and the second correlation model, the expected gas pressure value of each gas supply node in the next period is calculated to generate the corresponding gas pressure demand; According to the air pressure requirement, adjusting the air pressure of each air supply node and / or the conduction state between each air supply node; Wherein, the gas supply node includes a bifurcation point where at least two gas supply branches are formed by branching from the same gas supply main route; Adjusting the gas pressure of each gas supply node includes at least one of adjusting the gas pressure of a gas storage tank connected to the gas supply node, adjusting the gas pressure of an associated gas supply node, and adjusting the local temperature of the gas supply network.

[0008] Through the above technical solution, the gas demand of each gas supply node in the future set time period can be predicted more accurately, and then the gas pressure of the gas storage tank or the related gas supply node can be adjusted in advance, or the connectivity between the gas supply nodes can be changed to compensate for the response delay caused by signal transmission, air compressor start and stop, etc., and reduce the pressure fluctuation at the target gas supply node. At the same time, when adjusting the gas pressure of the gas supply node, the short-term effect of the gas temperature in the gas supply network on the gas pressure in the network is fully considered, and the pressure change caused by the temperature change of the local network is used to compensate for the response time required for the pressure to be transmitted along the gas supply network, thereby further shortening the gas supply response time at the target gas supply node, reducing the pressure fluctuation at the target gas supply node, achieving stable gas supply at the end of the gas supply network, and improving the gas supply quality.

[0009] Furthermore, the configuration method of each gas supply node includes: The gas supply network forms a tree-shaped gas supply structure from the gas storage tank toward the pneumatic terminal, and the gas supply nodes are divided into homologous nodes and heterologous nodes according to whether the gas storage tanks connected to the gas supply nodes are the same; Estimate the gas demand period of each pneumatic terminal at the end of the gas supply network according to the production line process, divide the pneumatic terminals with similar gas demand periods into one group to form multiple period groups; divide the pneumatic terminals located in the same area into one group according to the location layout of each pneumatic terminal to form multiple location groups; Select pneumatic terminals in the same position group and in different time groups, and connect them to share the same air supply node; Establishing controllable conduction connections between a number of homologous nodes and between a number of homologous nodes and heterologous nodes in a gas supply network and setting trigger conditions, and when the trigger conditions are met, conducting the number of homologous nodes, or the homologous nodes and heterologous nodes; The air pressure fluctuation correlation relationship between each air supply node in the first correlation model includes the correlation relationship between homologous nodes and the correlation relationship between homologous nodes and heterologous nodes in the on state; The first association model includes the relationship between the air pressure fluctuation amplitude and the fluctuation response delay between each air supply node.

[0010] Through the above technical solution, each pneumatic terminal can be evenly distributed to different gas supply nodes according to the location distribution of each pneumatic terminal and the gas demand time, avoiding the situation of concentrated gas use in the same period, reducing the air pressure fluctuation at the gas supply node, and ensuring the quality of gas supply. By establishing a controllable conduction relationship between homologous nodes and heterologous nodes, when the air pressure of multiple gas supply nodes connected to the same gas tank fluctuates greatly, the above air pressure fluctuation can be buffered by connecting other gas tanks, and the safety redundancy of the entire pneumatic system can also be improved.

[0011] Furthermore, a fluctuation model is generated to predict the fluctuation trend of the gas pressure at each gas supply node, including: Obtain the gas demand of each pneumatic terminal connected to the gas supply node in each time period, and integrate them to generate a theoretical gas consumption model; Obtain historical air pressure data and temperature data of the gas supply node, combine with the theoretical gas usage model, generate a correction function based on a neural network algorithm to correct the theoretical gas usage model, and generate the fluctuation model; The parameters of the correction function include time data, temperature data, and data on the conduction status between the current gas supply node and other gas supply nodes.

[0012] Through the above technical solution, a theoretical gas consumption model can be accurately generated based on the gas demand information of the pneumatic terminals connected to each gas supply node, and then the external influences such as temperature and other gas supply nodes can be incorporated into it, so that the generated fluctuation model can more accurately predict the air pressure fluctuation trend of the corresponding gas supply node.

[0013] Furthermore, according to the gas pressure requirement, the conduction state between the gas supply nodes is adjusted, including: Based on the fluctuation model corresponding to the current gas supply node, obtain the gas pressure fluctuation value P1 of the gas supply node in the next period; Obtain the conduction state between the current gas supply node and the associated gas supply node, calculate the air pressure fluctuation value P2 generated by the current gas supply node in the next time period under the influence of the connected associated gas supply node according to the first association model, and generate the time delay t1 required for the above air pressure fluctuation value P2 to act on the current gas supply node; According to the air pressure fluctuation value P1 and the air pressure fluctuation value P2, the theoretical air pressure value P3 of the current air supply node at each moment in the next period is obtained; Get the current air pressure value P0 of the gas supply node, and subtract it from the theoretical air pressure value P3 to get the air pressure value Pt that needs to be compensated; Obtaining the air pressure values ​​of other air supply nodes associated with the current air supply node, and adjusting the conduction state between the current air supply node and the associated air supply nodes according to the air pressure value Pt and the first association model; Wherein, adjusting the conduction state between the current gas supply node and the associated gas supply node includes: The target gas supply node ID that needs to be turned on and the time delay t2 required for the gas pressure to act on the current gas supply node after the target gas supply node is turned on are obtained from the first association model and the required gas pressure value Pt; Based on the time delay t2 and the time delay t1, the conduction time between the current gas supply node and the target gas supply node is determined, and based on the fluctuation model corresponding to the current gas supply node, the conduction duration between the current gas supply node and the target gas supply node is calculated and determined.

[0014] Through the above technical scheme, the fluctuation model corresponding to each gas supply node and the first association model can be used to predict the air pressure value that needs to be compensated for the current gas supply node in the next time period, and then the corresponding target gas supply node can be matched through the first association model, and the current gas supply node and the target gas supply node can be connected at the appropriate time. In this way, the associated gas supply nodes in the gas supply network can be used to achieve balanced distribution of gas pressure in the network and reduce air pressure fluctuations at the gas supply nodes.

[0015] Furthermore, the method of adjusting the local temperature of the gas supply network to adjust the gas pressure of the gas supply node includes: A heating device for heating the airflow in the gas supply network is arranged upstream of each gas supply node; Based on the correlation between the air flow temperature in a specific section of the air supply network and the air pressure at each air supply node stored in the second correlation model, directly generate the required regulated temperature T and the corresponding heating power and heating duration; and / or Based on the pipeline parameters and gas property parameters at the gas supply node, a corresponding heating calculation formula is configured for each gas supply node and stored in association with each gas supply node ID; Obtain the air pressure value Pt that needs to be compensated for the air supply node, and calculate the corresponding heating power and heating time according to the heating calculation formula.

[0016] Through the above technical solution, a heating device is directly configured at a position adjacent to the upstream of the gas supply node, and the heating parameters required to compensate for the air pressure value Pt are directly obtained through the second association model, which is convenient and efficient. When the second association model cannot quickly find the corresponding heating parameters under specific circumstances, the heating calculation formula corresponding to the gas supply node is used to quickly calculate the heating parameters required by the heating device, thereby responding to the air pressure demand at the gas supply node in a short time and reducing the air pressure fluctuation at the gas supply node.

[0017] Furthermore, the reliability of each volatility model is verified and a weight value is assigned to each volatility model, including: Import the historical air pressure data and temperature data of each gas supply node into the corresponding fluctuation model in units of time periods to obtain the expected air pressure value generated by the fluctuation model; Compare the expected air pressure value with the actual air pressure value in the historical air pressure data to generate air pressure deviation values ​​for each time period, and integrate the air pressure deviation values ​​based on the time period to form a deviation curve; The amplitude of the deviation curve of each gas supply node is counted, the maximum amplitude is obtained and the maximum amplitude is divided into multiple marked intervals with equal span; Associating multiple labeled intervals with specific assignments; The marking interval is used to calibrate the amplitude of each time period in each deviation curve and assign a value to each time period to form a weight value of the fluctuation model.

[0018] Through the above technical solution, a reliability weight value that fluctuates over time can be assigned to the fluctuation model corresponding to each gas supply node. This makes the prediction result more accurate when estimating the current gas supply node and calculating the impact of the associated gas supply node on the current gas supply node, which is beneficial to improving the accuracy of air pressure control and reducing air pressure fluctuations at the gas supply node.

[0019] Furthermore, based on the historical air pressure data and temperature data of each air supply node, a fluctuation model for predicting the air pressure fluctuation trend of each air supply node is generated, which also includes: Collect and store the air pressure data and temperature data of each gas supply node in a data storage unit, and establish a data connection between the data storage unit and the cloud server; Set the update time of each wave model; Detect whether the current time meets the set conditions. If so, send the data stored in the data storage unit and the current fluctuation model to the cloud server, generate a new correction function and correct the current fluctuation model; The modified fluctuation model is sent down and stored in association with the gas supply node ID.

[0020] Through the above technical solution, the cloud-based artificial intelligence big model can be used to regularly correct the fluctuation model of each gas supply node, making the prediction of the air pressure fluctuation trend at the gas supply node more accurate.

[0021] In order to implement the above-mentioned intelligent centralized control method of air compressor units, the present application also proposes an intelligent centralized control system of air compressor units, including a pipe network component and a control component; The pipe network assembly includes: at least two air compressors and a gas supply pipe network connected thereto, each gas flow distribution point in the gas supply pipe network forms a gas supply node, and a conducting pipe and a switch valve are arranged between specific gas supply nodes in the gas supply pipe network; Each of the gas supply nodes is provided with a temperature sensor for detecting the temperature of the gas flow and a pressure sensor for detecting the pressure at the gas supply node; a heating device for heating the gas flow in the gas supply network is provided at the upstream position of each of the gas supply nodes; The control component comprises: A data acquisition unit, configured to be connected to the temperature sensor, the pressure sensor and the switch valve signals, for acquiring real-time air pressure data, temperature data at each air supply node and conduction status data between each air supply node; A data storage unit, configured to be connected to the data acquisition unit and the cloud server data, and used to store historical air pressure data, temperature data and conduction status data of each gas supply node, a first correlation model used to characterize the correlation between air pressure fluctuations of each gas supply node, a second correlation model used to characterize the correlation between the air flow temperature of a specific section in the gas supply network and the air pressure of each gas supply node, conduction status data between each gas supply node in the gas supply network, and a fluctuation model corresponding to each gas supply node and its weight value; The data processing unit is configured to be data-connected with the data acquisition unit and the data storage unit, and is used to obtain the air pressure data, temperature data of each air supply node at the current moment, and the conduction state between each air supply node, and calculate the expected air pressure value of each air supply node in the next period according to the fluctuation model and weight value of each air supply node, combined with the first correlation model and the second correlation model, to generate the corresponding air pressure demand; A control output unit is configured to be data-connected to the data processing unit and control-connected to the air compressor and each of the heating devices and the switch valve, and to adjust the air pressure of each air supply node and / or the conduction state between each air supply node according to the air pressure demand; Among them, adjusting the air pressure of each gas supply node includes: adjusting the air pressure of the gas tank connected to the gas supply node by adjusting the operating state of the air compressor, adjusting the air pressure of the gas supply node associated with the current gas supply node, and adjusting at least one of the local temperature of the gas supply network.

[0022] Furthermore, the control component also includes: A fluctuation model weight configuration unit, configured to import the historical air pressure data and temperature data of each gas supply node into the corresponding fluctuation model in units of time periods, verify the reliability of each fluctuation model, assign a weight value to each fluctuation model, and output it to the data processing unit; The fluctuation model updating unit is configured to collect and store the air pressure data and temperature data of each gas supply node in the data storage unit based on the set update time node, and send the data stored in the data storage unit and the current fluctuation model to the cloud server, generate a new correction function and correct the current fluctuation model, and send the corrected fluctuation model to the data storage unit, and store it in association with the gas supply node ID.

[0023] A computer-readable storage medium for intelligent centralized control of an air compressor unit, wherein a computer program is stored on the storage medium, and when the computer program is executed by a processor, the steps of the intelligent centralized control method of an air compressor unit as described above are implemented.

[0024] Compared with the prior art, the present invention has the following beneficial effects: (1) By collecting the historical fluctuation data of each gas supply node, a fluctuation model is established for each gas supply node, and a weight value representing the reliability is configured for each fluctuation model. This can predict the pressure fluctuation trend of each gas supply node in the future set period. At the same time, by establishing the correlation between the pressure fluctuations of each gas supply node, the pressure change of a single gas supply node can be predicted more accurately, so that the control system can make control adjustments in advance, shorten the response time of the entire pneumatic system, reduce the pressure fluctuation at the gas supply node, and improve the gas supply quality. (2) By regulating the temperature of the local gas supply network, the air pressure at the local location in the gas supply network can be accurately adjusted to compensate for the time delay of the air pressure being transmitted from the gas storage tank / air compressor outlet to the gas supply node, shorten the gas demand response time at the end of the gas supply network, and reduce the air pressure fluctuation at the gas supply node and pneumatic terminal; (3) By connecting the gas supply nodes belonging to different air compressors / gas tanks in the gas supply network at the appropriate time, the entire pneumatic system can be prevented from being shut down due to the failure of a single gas tank / air compressor, thereby improving the safety redundancy of the entire pneumatic system. At the same time, the connection relationship between heterogeneous nodes can be used to balance the air pressure in the entire gas supply network and reduce the air pressure fluctuation in the gas supply network. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is an overall schematic diagram of the method of the present invention; Figure 2 Schematic diagram of the method for generating a volatility model; Figure 3 Schematic diagram of the method for assigning weights to the volatility model. DETAILED DESCRIPTION

[0026] The present application is further described in detail below in conjunction with embodiments and drawings, but the implementation methods of the present application are not limited thereto.

[0027] In order to make the description of the implementation mode of the present application clearer, before further describing the intelligent centralized control method of the air compressor unit in detail, the relevant feature names involved in the implementation mode of the present application are defined, and the relevant structural configuration of the gas supply network is briefly described: The gas supply network involved in this application adopts a common tree-shaped gas supply structure, that is, the gas supply network extends in a tree-shaped layout from the gas storage tank toward the pneumatic terminal. The specific configuration order is: air compressor, gas storage tank, filter, dryer, gas supply pipeline, and gas ports connected to each pneumatic terminal.

[0028] In the implementation manner of the present application, the air supply node is defined as a bifurcation point where at least two air supply branches are formed by diverting from the same air supply main line. For example, if the air outlet of the dryer is connected to multiple air supply branches through an air supply main pipe, it can be regarded as an air supply node here. The simplest air supply node is a "Y"-shaped node.

[0029] At the same time, in the gas supply network with multiple gas tanks, each gas supply node is divided into a homologous node and a heterologous node according to whether the gas tanks to which each gas supply node is ultimately connected are the same.

[0030] Based on the above definitions and descriptions, in order to more clearly illustrate and implement the intelligent centralized control method of the air compressor unit described in this application, in the implementation mode of this application, the configuration method of each of the air supply nodes is as follows: According to the production line process, the gas demand period of each pneumatic terminal at the end of the gas supply network is estimated, and the pneumatic terminals with similar gas demand periods are divided into a group to form multiple time period groups; according to the location layout of each pneumatic terminal, the pneumatic terminals in the same area are divided into a group to form multiple location groups. Then, the pneumatic terminals in the same location group and in different time period groups are selected to connect and share the same gas supply node.

[0031] The above technical solution can evenly distribute each pneumatic terminal to different gas supply nodes according to its location distribution and gas demand time, avoid concentrated gas consumption in the same period, reduce air pressure fluctuations at the gas supply nodes, and ensure gas supply quality.

[0032] In order to improve the safety redundancy of the pneumatic system, and in special cases, heterogeneous nodes can be used to buffer the pressure fluctuations of adjacent gas supply nodes, when configuring the gas supply nodes, it also includes: establishing controllable conductive connections between several homogeneous nodes in the gas supply network, and between several homogeneous nodes and heterogeneous nodes, and setting trigger conditions. When the trigger conditions are met, several homogeneous nodes, or homogeneous nodes and heterogeneous nodes are connected. Therefore, when the air pressure of multiple gas supply nodes connected to the same gas tank fluctuates greatly, the above pressure fluctuations can be buffered by connecting other gas tanks.

[0033] The intelligent centralized control method of an air compressor unit described in the embodiment of the present application is based on the air compressor unit and the gas supply network connected thereto, such as Figure 1 As shown, it mainly includes the following steps: S100, establishing and storing a first correlation model for characterizing the correlation relationship between air pressure fluctuations of each air supply node; S200, establishing and storing a second correlation model for characterizing the correlation between the air flow temperature in a specific section of the gas supply network and the air pressure of each gas supply node; S300, acquiring and generating a fluctuation model for predicting the pressure fluctuation trend of each gas supply node based on the historical pressure data and temperature data of each gas supply node, and storing each fluctuation model in association with each gas supply node ID; S400, based on the historical air pressure data and temperature data of each gas supply node, verify the reliability of each fluctuation model and assign a weight value to each fluctuation model; S500, obtaining the air pressure data and temperature data of each air supply node at the current moment and the conduction status between each air supply node, and calculating the expected air pressure value of each air supply node in the next period according to the fluctuation model and weight value of each air supply node, combined with the first association model and the second association model, to generate the corresponding air pressure demand; S600, adjusting the air pressure of each air supply node and / or the conduction state between each air supply node according to the air pressure demand. In the above step S600, adjusting the air pressure of each air supply node includes at least one of adjusting the air pressure of the air storage tank connected to the air supply node, adjusting the air pressure of the associated air supply node, and adjusting the local temperature of the air supply network.

[0034] In the above step S100, the pressure fluctuation correlation between each gas supply node in the first correlation model includes the correlation between homologous nodes and the correlation between homologous nodes and heterologous nodes in the on state. The above correlation includes the pressure fluctuation amplitude relationship and the fluctuation response delay relationship between each gas supply node.

[0035] As described in detail, since the nodes of the same source are interconnected, when the air pressure of one of the gas supply nodes drops, it will inevitably cause the air pressure of the adjacent or even farther gas supply nodes to fluctuate. Usually, the law of the correlation fluctuation of the above gas supply nodes can be obtained in a closed gas supply network. The way to obtain the above correlation relationship includes but is not limited to: obtaining it through theoretical deduction based on the physical connection structure of the gas supply network; or collecting the air pressure data of each gas supply node during the system test and obtaining it through a fitting algorithm; or using a neural network model and obtaining it through big data training.

[0036] The correlation relationship in the first correlation model includes the amplitude relationship and the fluctuation response delay. For example, when only considering the correlation relationship between the two gas supply nodes A and B, when the air pressure at the gas supply node A fluctuates, the air pressure at the adjacent gas supply node B fluctuates accordingly, and the correlation relationship between the two can be defined as: the amplitude of the fluctuation of the gas supply node B following the gas supply node A, and the response time of the gas supply node B.

[0037] In the above step S200, a second correlation model is established and stored to characterize the correlation between the air flow temperature in a specific section of the gas supply network and the air pressure of each gas supply node. The above second correlation model refers to a specific section in the gas supply network, such as the temperature in a set length of pipeline upstream of a gas supply node, or the temperature of a specific gas supply node. In a closed space, the temperature and pressure of the gas are linearly related. Therefore, the above second correlation model can be a directly stored temperature-pressure data table, or a calculation formula model for calculating the air pressure based on the current gas temperature.

[0038] In the implementation manner of the present application, in order to facilitate rapid acquisition of the impact results of the associated gas supply nodes, the first association model and the second association model are preferably stored in the form of a two-dimensional data table.

[0039] In the above step S300, a fluctuation model is generated to predict the pressure fluctuation trend of each gas supply node, such as Figure 2 As shown, including: S310, obtain the gas demand of each pneumatic terminal connected to the gas supply node in each time period, and integrate them to generate a theoretical gas consumption model. In detail, the above gas demand can be obtained according to the type of pneumatic terminal and the steps of the production process. For example, if the pneumatic terminal is a cylinder, the amount of gas consumed by the cylinder each time it moves is fixed, and on the automated production line, the operating time of each pneumatic terminal is often fixed. Therefore, the gas demand of each gas supply node in a certain period of time can be calculated. In practice, the gas demand of the gas supply node in each time period can also be obtained based on the air pressure data statistics when the pneumatic system is tested.

[0040] The theoretical gas usage model obtained through theoretical calculation or test data will also be affected by many factors in specific production practice, such as the ambient temperature of the production line, temporary changes in the conduction state between gas supply nodes, etc. In order to more accurately predict the pressure fluctuation trend of each gas supply node in the next period of time, the generation of the fluctuation model described in this application also includes the following steps: S320, obtaining historical air pressure data and temperature data of the gas supply node, combining the theoretical gas usage model, and generating a correction function based on a neural network algorithm to correct the theoretical gas usage model to generate the fluctuation model. The parameter variables of the correction function include time data, temperature data, and the conduction state data between the current gas supply node and other gas supply nodes.

[0041] It should be pointed out that the generated fluctuation model is actually a fluctuation curve whose parameters are time and temperature, which reflects the changes in the air pressure value of the gas supply node under different time and temperature conditions.

[0042] The above-mentioned scheme for generating the fluctuation model combines the theoretical model with practical application, and incorporates external influences such as temperature and other gas supply nodes, so that the generated fluctuation model can more accurately predict the pressure fluctuation trend of the corresponding gas supply node.

[0043] In order to facilitate the processing and prediction of the air pressure data of each gas supply node, in the implementation mode of the present application, the above-mentioned fluctuation model is stored in association with the gas supply node ID information. In the working state, the corresponding fluctuation model can be quickly retrieved by simply inputting the ID of the gas supply node.

[0044] In order to ensure that each gas supply node can accurately supply gas and shorten the response time of each gas supply node, the fluctuation model used to predict the gas pressure fluctuation trend of each gas supply node is generated based on the historical gas pressure data and temperature data of each gas supply node, and the waveform automatic update optimization step is also included: S331, collect and store the air pressure data and temperature data of each gas supply node in the data storage unit, establish a data connection between the above data storage unit and the cloud server. In actual applications, the data at each gas supply node is summarized in the factory or workshop server, and then uploaded to the cloud through the network.

[0045] S332, setting the update time of each fluctuation model.

[0046] S333, detect whether the current time meets the set conditions. If so, send the data stored in the data storage unit and the current fluctuation model to the cloud server, generate a new correction function and correct the current fluctuation model.

[0047] S334, issuing the modified fluctuation model and storing it in association with the gas supply node ID.

[0048] The above technical solution can utilize the cloud-based artificial intelligence large model provided by the data service company to regularly correct the fluctuation model of each gas supply node, making the prediction of the air pressure fluctuation trend at the gas supply node more accurate, while also reducing the computing power configuration at the production site.

[0049] In actual applications, since there may be a conductive connection between the gas supply nodes, when the air pressure of one of the gas supply nodes fluctuates, the fluctuation will be transmitted to other gas supply nodes. In order to reduce the impact of high-fluctuation gas supply nodes on the entire gas supply network, in the implementation of this application, the reliability of each gas supply node will also be verified.

[0050] In step S400, the reliability of each fluctuation model is verified and a weight value is assigned to each fluctuation model. Figure 3 As shown, further comprising: S410, importing historical air pressure data and temperature data of each gas supply node into a corresponding fluctuation model in units of time periods, and obtaining an expected air pressure value generated by the fluctuation model.

[0051] S420, comparing the expected air pressure value with the actual air pressure value in the historical air pressure data, generating air pressure deviation values ​​for each time period, and integrating the air pressure deviation values ​​based on the time period to form a deviation curve.

[0052] S430, count the amplitudes of the deviation curves of each gas supply node, obtain the maximum amplitude and divide it into multiple marked intervals with equal span. In practical applications, in order to eliminate interference items, the maximum and minimum amplitude data can be removed. The marked interval means dividing the amplitude into multiple numerical intervals, and the amplitudes falling into the same numerical interval will be marked with the same numerical value, that is, assigned a specific value.

[0053] S440, associating multiple tag intervals with specific assignments.

[0054] S450, using the marking interval to calibrate the amplitude of each time period in each deviation curve and assign a value to each time period to form a weight value of the fluctuation model.

[0055] The above step S400 can assign a reliability weight value that fluctuates over time to the fluctuation model corresponding to each gas supply node. For example, if there are three associated gas supply nodes, A, B, and C, according to the first associated model, when the air pressure of the two gas supply nodes A and B decreases by 0.1MPa respectively, the air pressure of the gas supply node C decreases by 0.08MPa and 0.06MPa respectively. However, since the reliability weight value of the fluctuation model of the gas supply node A is 1 and the reliability weight value of the fluctuation model of the gas supply node B is 0.8, when the air pressure of the two gas supply nodes A and B decreases by 0.1MPa at the same time, the air pressure of the gas supply node C decreases by 0.08*1+0.06*0.8=0.128MPa. It can be seen from the above estimation process that when estimating the current gas supply node and calculating the impact of the associated gas supply node on the current gas supply node, introducing the reliability weight value can make the prediction result more accurate, which is beneficial to improving the accuracy of gas pressure control and reducing the pressure fluctuation at the gas supply node.

[0056] In step S600, according to the gas pressure requirement, the conduction state between the gas supply nodes is adjusted, further comprising: S610, based on the fluctuation model corresponding to the current gas supply node, obtain the air pressure fluctuation value P1 of the gas supply node in the next period. When the duration of the next period is short, the air pressure fluctuation value P1 can be regarded as a fixed value, and when the duration of the next period is long, the air pressure fluctuation value P1 can be regarded as a fluctuation curve with a period of time as the horizontal axis and a fluctuation amplitude as the vertical axis.

[0057] S611, obtaining the conduction state between the current gas supply node and the associated gas supply node, calculating the air pressure fluctuation value P2 of the current gas supply node in the next period of time caused by the connected associated gas supply node according to the first association model, and generating the time delay t1 required for the air pressure fluctuation value P2 to act on the current gas supply node; S612, according to the air pressure fluctuation value P1 and the air pressure fluctuation value P2, obtain the theoretical air pressure value P3 of the current air supply node at each moment in the next period. Similarly, the theoretical air pressure value P3 can be a fixed value or a fluctuation curve.

[0058] S613, obtaining the current air pressure value P0 of the air supply node, and subtracting it from the theoretical air pressure value P3 to obtain the air pressure value Pt that needs to be compensated; S614, obtaining the air pressure values ​​of other air supply nodes associated with the current air supply node, and adjusting the conduction state between the current air supply node and the associated air supply nodes according to the air pressure value Pt and the first association model. In practical applications, it is only necessary to find a match or calculate based on the first association model to achieve the conduction state combination of the air pressure value Pt.

[0059] In the above step S650, in detail, in one embodiment, adjusting the conduction state between the current gas supply node and the associated gas supply node includes: S6141, obtaining the target gas supply node ID that needs to be turned on and the time delay t2 required for the gas pressure to act on the current gas supply node after the target gas supply node is turned on, based on the first association model and the required gas pressure value Pt; S6142, based on the time delay t2 and the time delay t1, determine the conduction time between the current gas supply node and the target gas supply node, and calculate and determine the conduction duration between the current gas supply node and the target gas supply node based on the fluctuation model corresponding to the current gas supply node.

[0060] The above technical solution can predict the air pressure value that needs to be compensated for the current gas supply node in the next time period through the fluctuation model corresponding to each gas supply node and the first association model, and then obtain the corresponding target gas supply node through matching through the first association model, and connect the current gas supply node and the target gas supply node at the appropriate time. In this way, the associated gas supply nodes in the gas supply network can be used to achieve balanced distribution of gas pressure in the gas supply network and reduce air pressure fluctuations at the gas supply nodes.

[0061] In step S600, the method of adjusting the local temperature of the gas supply network to adjust the gas pressure of the gas supply node specifically includes: S620, a heating device, such as a heating wire or a heating plate, is configured upstream of each gas supply node for heating the gas flow in the gas supply network.

[0062] S6211, directly generating the required regulated temperature and the corresponding heating power and heating duration based on the correlation between the air flow temperature in a specific section of the air supply network and the air pressure of each air supply node stored in the second correlation model; and / or S6212: Based on the pipeline parameters and gas property parameters at the gas supply node, a corresponding heating calculation formula is configured for each gas supply node and stored in association with each gas supply node ID.

[0063] S622, obtaining the gas pressure value Pt that needs to be compensated for the gas supply node, and calculating the corresponding heating power and heating time according to the gas heating calculation formula.

[0064] In the above technical solution, a heating device is directly configured at a position adjacent to the upstream of the gas supply node, and the heating parameters required to compensate for the gas pressure value Pt are directly obtained through the second association model, which is convenient and efficient. When the second association model cannot quickly find the corresponding heating parameters under certain circumstances, the heating parameters required by the heating device are quickly calculated through the heating calculation formula corresponding to the gas supply node, thereby responding to the gas pressure demand at the gas supply node in a short time and reducing the gas pressure fluctuation at the gas supply node.

[0065] In order to implement the above-mentioned intelligent centralized control method of the air compressor unit, the present application also proposes an intelligent centralized control system of the air compressor unit, which mainly includes a pipe network component and a control component.

[0066] The pipe network assembly includes at least two air compressors and a gas supply network connected thereto. Each gas flow distribution point in the gas supply network forms a gas supply node. A conducting pipeline and a switch valve are arranged between specific gas supply nodes in the gas supply network. The switch valve is configured as a solenoid valve electrically connected to the control assembly for easy switch control.

[0067] Each of the gas supply nodes is provided with a temperature sensor for detecting the temperature of the air flow and a pressure sensor for detecting the pressure at the gas supply node. In a specific embodiment, an air flow sensor can be used to reversely calculate the air pressure of the gas supply node based on the detected air flow data. A heating device for heating the air flow in the gas supply network is provided at the upstream position of each of the gas supply nodes. The heating device is preferably configured as an electric heater arranged side by side in the gas supply pipeline along the air flow direction, and the control end of the electric heater is connected to the control component. In actual applications, the installation position of the electric heater is located upstream of the gas supply node, and the distance between the electric heater and the gas supply node is determined by the air flow velocity in the gas supply pipeline and the heating power of the electric heater, so as to ensure that the heated gas can reach the gas supply node within the set time period.

[0068] In the implementation manner of the present application, the control component mainly includes a data acquisition unit, a data storage unit, a data processing unit, and a control output unit.

[0069] The data acquisition unit is configured to be connected to the temperature sensor, the pressure sensor and the signals of each switch valve, and is used to obtain the real-time air pressure data, temperature data at each air supply node and the conduction status between each air supply node.

[0070] The data storage unit is configured to be data-connected to the data acquisition unit and the cloud server, and can be implemented by a data storage device and a network communication module arranged inside a factory or workshop.

[0071] The data storage unit is mainly used to store the historical air pressure data and temperature data of each gas supply node, the first correlation model used to characterize the correlation between the air pressure fluctuations of each gas supply node, the second correlation model used to characterize the correlation between the air flow temperature of a specific section in the gas supply network and the air pressure of each gas supply node, the conduction state data between each gas supply node in the gas supply network, and the fluctuation model and its weight value corresponding to each gas supply node. Preferably, all the above data are synchronized to the cloud storage in real time through the network.

[0072] The data processing unit is configured to be data-connected with the data acquisition unit and the data storage unit, and is used to obtain the air pressure data, temperature data of each air supply node and the conduction status between each air supply node at the current moment. The expected air pressure value of each air supply node in the next time period is calculated and / or searched based on the fluctuation model and weight value of each air supply node, combined with the first correlation model and the second correlation model, to generate the corresponding air pressure demand of each air supply node. In practical applications, the above-mentioned data processing unit can be completed by a PC configured at the production site.

[0073] The control output unit is configured to be data-connected to the data processing unit, and to be control-connected to the air compressor and each of the heating devices and the switch valves. According to the air pressure demand, in combination with the first association model and the second association model, a corresponding control command signal is generated and sent to the air compressor, each of the switch valves and the heating device to adjust the air pressure of each air supply node and / or the conduction state between each air supply node. Among them, adjusting the air pressure of each air supply node includes: adjusting the air pressure of the air tank connected to the air supply node by adjusting the operating state of the air compressor, adjusting the air pressure of the air supply node associated with the current air supply node, and adjusting at least one of the local temperature of the air supply network. In practical applications, the above-mentioned control output unit can be configured as a PLC control module connected to a host computer.

[0074] In order to update the volatility model and improve the prediction accuracy of each volatility model, the control component also includes a volatility model weight configuration unit and a volatility model update unit.

[0075] The fluctuation model weight configuration unit is configured to import the historical air pressure data and temperature data of each gas supply node into the corresponding fluctuation model in units of time periods, verify the reliability of each fluctuation model, assign a weight value to each fluctuation model and output it to the data processing unit. The specific implementation method is as described in the previous method and will not be repeated here.

[0076] The fluctuation model update unit is configured to collect and store the air pressure data and temperature data of each gas supply node in the data storage unit based on the set update time node, and send the data stored in the data storage unit and the current fluctuation model to the cloud server, generate a new correction function and correct the current fluctuation model, and send the corrected fluctuation model to the data storage unit, and store it in association with the gas supply node ID.

[0077] Finally, in order to facilitate the promotion and use of the intelligent centralized control method for air compressor units described in the implementation mode of this application, it is also proposed to protect a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the intelligent centralized control method for air compressor units as described above are implemented. The above-mentioned computer-readable storage medium includes but is not limited to an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the above.

[0078] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. An intelligent centralized control method for an air compressor unit, based on an air compressor unit and a connected air supply network, characterized in that: include: Establishing and storing a first correlation model for characterizing the correlation relationship between the air pressure fluctuations of each air supply node; Establishing and storing a second correlation model for characterizing the correlation between the air flow temperature in a specific section of the air supply network and the air pressure at each air supply node; Obtain and generate a fluctuation model for predicting the pressure fluctuation trend of each gas supply node based on the historical air pressure data and temperature data of each gas supply node, and store each fluctuation model in association with the ID of each gas supply node; Based on the historical air pressure and temperature data of each gas supply node, the reliability of each fluctuation model is verified and a weight value is assigned to each fluctuation model; Obtain the air pressure data, temperature data of each air supply node at the current moment and the conduction status between each air supply node; According to the fluctuation model and weight value of each gas supply node, combined with the first correlation model and the second correlation model, the expected gas pressure value of each gas supply node in the next period is calculated to generate the corresponding gas pressure demand; According to the air pressure requirement, adjusting the air pressure of each air supply node and / or the conduction state between each air supply node; Wherein, the gas supply node includes a bifurcation point where at least two gas supply branches are formed by branching from the same gas supply main route; Adjusting the gas pressure of each gas supply node includes at least one of adjusting the gas pressure of a gas storage tank connected to the gas supply node, adjusting the gas pressure of an associated gas supply node, and adjusting the local temperature of the gas supply network.

2. The intelligent centralized control method for air compressor units according to claim 1 is characterized in that: The configuration method of each gas supply node includes: The gas supply network forms a tree-shaped gas supply structure from the gas storage tank toward the pneumatic terminal, and the gas supply nodes are divided into homologous nodes and heterologous nodes according to whether the gas storage tanks connected to the gas supply nodes are the same; Estimate the gas demand period of each pneumatic terminal at the end of the gas supply network according to the production line process, divide the pneumatic terminals with similar gas demand periods into one group to form multiple period groups; divide the pneumatic terminals located in the same area into one group according to the location layout of each pneumatic terminal to form multiple location groups; Select pneumatic terminals in the same position group and in different time groups, and connect them to share the same air supply node; Establishing controllable conduction connections between a number of homologous nodes and between a number of homologous nodes and heterologous nodes in a gas supply network and setting trigger conditions, and when the trigger conditions are met, conducting the number of homologous nodes, or the homologous nodes and heterologous nodes; The air pressure fluctuation correlation relationship between each air supply node in the first correlation model includes the correlation relationship between homologous nodes and the correlation relationship between homologous nodes and heterologous nodes in the on state; The first association model includes the relationship between the air pressure fluctuation amplitude and the fluctuation response delay between each air supply node.

3. The intelligent centralized control method for air compressor units according to claim 2 is characterized in that: Generate a fluctuation model to predict the pressure fluctuation trend of each gas supply node, including: Obtain the gas demand of each pneumatic terminal connected to the gas supply node in each time period, and integrate them to generate a theoretical gas consumption model; Obtain historical air pressure data and temperature data of the gas supply node, combine with the theoretical gas usage model, generate a correction function based on a neural network algorithm to correct the theoretical gas usage model, and generate the fluctuation model; The parameters of the correction function include time data, temperature data, and data on the conduction status between the current gas supply node and other gas supply nodes.

4. The intelligent centralized control method for air compressor units according to claim 1, characterized in that: According to the gas pressure requirements, adjust the conduction status between each gas supply node, including: Based on the fluctuation model corresponding to the current gas supply node, obtain the gas pressure fluctuation value P1 of the gas supply node in the next period; Obtain the conduction state between the current gas supply node and the associated gas supply node, calculate the air pressure fluctuation value P2 generated by the current gas supply node in the next time period under the influence of the connected associated gas supply node according to the first association model, and generate the time delay t1 required for the above air pressure fluctuation value P2 to act on the current gas supply node; According to the air pressure fluctuation value P1 and the air pressure fluctuation value P2, the theoretical air pressure value P3 of the current air supply node at each moment in the next period is obtained; Get the current air pressure value P0 of the gas supply node, and subtract it from the theoretical air pressure value P3 to get the air pressure value Pt that needs to be compensated; Obtaining the air pressure values ​​of other air supply nodes associated with the current air supply node, and adjusting the conduction state between the current air supply node and the associated air supply nodes according to the air pressure value Pt and the first association model; Wherein, adjusting the conduction state between the current gas supply node and the associated gas supply node includes: The target gas supply node ID that needs to be turned on and the time delay t2 required for the gas pressure to act on the current gas supply node after the target gas supply node is turned on are obtained from the first association model and the required gas pressure value Pt; Based on the time delay t2 and the time delay t1, the conduction time between the current gas supply node and the target gas supply node is determined, and based on the fluctuation model corresponding to the current gas supply node, the conduction duration between the current gas supply node and the target gas supply node is calculated and determined.

5. The intelligent centralized control method for air compressor units according to claim 4 is characterized in that: Methods for adjusting the local temperature of the gas supply network to adjust the gas pressure at the gas supply node include: A heating device for heating the airflow in the gas supply network is arranged upstream of each gas supply node; Based on the correlation between the air flow temperature in a specific section of the air supply network and the air pressure at each air supply node stored in the second correlation model, directly generate the required regulated temperature T and the corresponding heating power and heating duration; and / or Based on the pipeline parameters and gas property parameters at the gas supply node, a corresponding heating calculation formula is configured for each gas supply node and stored in association with each gas supply node ID; Obtain the air pressure value Pt that needs to be compensated for the air supply node, and calculate the corresponding heating power and heating time according to the heating calculation formula.

6. The intelligent centralized control method for air compressor units according to claim 1, characterized in that: Verify the reliability of each volatility model and assign weight values ​​to each volatility model, including: Import the historical air pressure data and temperature data of each gas supply node into the corresponding fluctuation model in units of time periods to obtain the expected air pressure value generated by the fluctuation model; Compare the expected air pressure value with the actual air pressure value in the historical air pressure data to generate air pressure deviation values ​​for each time period, and integrate the air pressure deviation values ​​based on the time period to form a deviation curve; The amplitude of the deviation curve of each gas supply node is counted, the maximum amplitude is obtained and the maximum amplitude is divided into multiple marked intervals with equal span; Associating multiple labeled intervals with specific assignments; The marking interval is used to calibrate the amplitude of each time period in each deviation curve and assign a value to each time period to form a weight value of the fluctuation model.

7. The intelligent centralized control method for air compressor units according to claim 6, characterized in that: Based on the historical air pressure data and temperature data of each air supply node, a fluctuation model is generated to predict the air pressure fluctuation trend of each air supply node, which also includes: Collect and store the air pressure data and temperature data of each gas supply node in a data storage unit, and establish a data connection between the data storage unit and the cloud server; Set the update time of each wave model; Detect whether the current time meets the set conditions. If so, send the data stored in the data storage unit and the current fluctuation model to the cloud server, generate a new correction function and correct the current fluctuation model; The modified fluctuation model is sent down and stored in association with the gas supply node ID.

8. An intelligent centralized control system for air compressor units, characterized in that: Including pipe network components and control components; The pipe network assembly includes: at least two air compressors and a gas supply pipe network connected thereto, each gas flow distribution point in the gas supply pipe network forms a gas supply node, and a conducting pipe and a switch valve are arranged between specific gas supply nodes in the gas supply pipe network; Each of the gas supply nodes is provided with a temperature sensor for detecting the temperature of the gas flow and a pressure sensor for detecting the pressure at the gas supply node; a heating device for heating the gas flow in the gas supply network is provided at the upstream position of each of the gas supply nodes; The control component comprises: A data acquisition unit, configured to be connected to the temperature sensor, the pressure sensor and the switch valve signals, for acquiring real-time air pressure data, temperature data at each air supply node and conduction status data between each air supply node; A data storage unit, configured to be connected to the data acquisition unit and the cloud server data, and used to store historical air pressure data, temperature data and conduction status data of each gas supply node, a first correlation model used to characterize the correlation between air pressure fluctuations of each gas supply node, a second correlation model used to characterize the correlation between the air flow temperature of a specific section in the gas supply network and the air pressure of each gas supply node, conduction status data between each gas supply node in the gas supply network, and a fluctuation model corresponding to each gas supply node and its weight value; The data processing unit is configured to be data-connected with the data acquisition unit and the data storage unit, and is used to obtain the air pressure data, temperature data of each air supply node at the current moment, and the conduction state between each air supply node, and calculate the expected air pressure value of each air supply node in the next period according to the fluctuation model and weight value of each air supply node, combined with the first correlation model and the second correlation model, to generate the corresponding air pressure demand; A control output unit is configured to be data-connected to the data processing unit and control-connected to the air compressor and each of the heating devices and the switch valve, and to adjust the air pressure of each air supply node and / or the conduction state between each air supply node according to the air pressure demand; Among them, adjusting the air pressure of each gas supply node includes: adjusting the air pressure of the gas tank connected to the gas supply node by adjusting the operating state of the air compressor, adjusting the air pressure of the gas supply node associated with the current gas supply node, and adjusting at least one of the local temperature of the gas supply network.

9. The intelligent centralized control system for air compressor units according to claim 8 is characterized in that: The control component also includes: A fluctuation model weight configuration unit, configured to import the historical air pressure data and temperature data of each gas supply node into the corresponding fluctuation model in units of time periods, verify the reliability of each fluctuation model, assign a weight value to each fluctuation model, and output it to the data processing unit; The fluctuation model updating unit is configured to collect and store the air pressure data and temperature data of each gas supply node in the data storage unit based on the set update time node, and send the data stored in the data storage unit and the current fluctuation model to the cloud server, generate a new correction function and correct the current fluctuation model, and send the corrected fluctuation model to the data storage unit, and store it in association with the gas supply node ID.

10. A computer-readable storage medium for intelligent centralized control of an air compressor unit, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the processor, the steps of the intelligent centralized control method for an air compressor unit as described in any one of claims 1 to 7 are implemented.

Images