A control method, system, device and storage medium of an air compressor unit

Abstract

The application discloses a kind of air compressor unit control method, equipment and computer readable storage medium, the method comprises: when detecting that the first gas parameter and the second gas parameter of the action environment of air compressor unit at least one exceeds preset range, with the second gas parameter of action environment at next time, state control information is obtained, for controlling the use state of each air compressor in target sub air compressor unit, to make the gas flow of target sub air compressor unit next time reach target gas flow, by the above-mentioned way, the second gas parameter obtained by prediction is used to control air compressor unit, compared with the control using current time gas parameter, control is carried out early, avoid the gas parameter still increases according to original trend in short time and leads to gas parameter over-limit, accurately and timely control is carried out to gas parameter, control the gas parameter of action environment in more accurate range, improve the accuracy and timeliness of air compressor unit control.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of air compressors, in particular to a control method and device for an air compressor unit and a computer readable storage medium. BACKGROUND

[0002] An air compressor, which can also be referred to as an air compressor unit, is a device for compressing gas. Multiple air compressors can be combined to form an air compressor unit, which can be used as a gas source for production to supply gas to a gas-consuming end. In order to meet the needs of production, the air compressor unit needs to be controlled so that the flow and pressure of the gas-consuming end meet the needs of production and do not cause waste.

[0003] In the prior art, the pressure and flow of the gas-consuming end are usually monitored to determine whether the pressure and flow of the gas-consuming end are out of limits, or the subsequent flow is predicted based on historical flow changes, so as to control the air compressor unit and achieve control of the pressure and flow of the gas-consuming end. The present inventors have found in a long-term research and development process that the existing control method for the air compressor unit is not accurate enough and cannot timely adjust the pressure and flow of the gas-consuming end within a more accurate range. SUMMARY

[0004] The technical problem solved by the present application is to provide a control method, device and computer readable storage medium for an air compressor unit, which can improve the accuracy and timeliness of air compressor unit control.

[0005] To solve the above technical problem, one technical solution adopted by the present application is to provide a control method for an air compressor unit, which comprises: obtaining a gas characteristic parameter of an action environment of the air compressor unit, wherein the gas characteristic parameter comprises a first gas parameter of the action environment at a current time and a second gas parameter at a next time; detecting that at least one of the first gas parameter and the second gas parameter exceeds a preset range; obtaining state control information of a target sub-air compressor unit using the second gas parameter, the target sub-air compressor unit comprising at least one air compressor in the air compressor unit; and controlling the use state of each air compressor in the target sub-air compressor unit based on the state control information.

[0006] To solve the above technical problem, another technical solution adopted by the present application is to provide an air compressor control system, which comprises an air compressor unit and a control device, wherein the control device executes the above control method for the air compressor unit to control the use state of at least one air compressor in the air compressor unit.

[0007] To solve the above technical problem, another technical solution adopted by the present application is to provide an electronic device, which comprises a processor and a memory, the memory being used to store program data, and the processor being used to execute the program data to implement the above control method for the air compressor unit.

[0008] To solve the above technical problems, another technical solution adopted by the present application is to provide a computer readable storage medium for storing program data, which can be executed to implement the control method of the air compressor unit.

[0009] In the above scheme, when it is detected that at least one of the first gas parameter and the second gas parameter of the working environment of the air compressor unit exceeds the preset range, the state control information of the target sub-air compressor unit is obtained by using the second gas parameter of the working environment at the next moment, and based on the state control information, the use state of each air compressor in the target sub-air compressor unit is controlled, so that the gas flow of the target sub-air compressor unit at the next moment reaches the target gas flow. By the above-mentioned method, the second gas parameter of the working environment at the next moment is predicted and used to control the air compressor unit. Compared with the control using the first gas parameter, the control is performed earlier, and the gas parameter is still increased according to the original trend in a short time, which avoids the gas parameter exceeding the limit. The gas parameter is accurately and timely controlled, the gas parameter of the working environment is controlled in a more accurate range, and the accuracy and timeliness of the air compressor unit control are improved. BRIEF DESCRIPTION OF DRAWINGS

[0010] Figure 1 is a flowchart of an embodiment of the control method of the air compressor unit of the present application;

[0011] Figure 2 is a flowchart of another embodiment of the control method of the air compressor unit of the present application;

[0012] Figure 3 is a flowchart of another embodiment of step S250 of the present application;

[0013] Figure 4 is a flowchart of another embodiment of step S351 of the present application;

[0014] Figure 5 is a flowchart of another embodiment of the control method of the air compressor unit of the present application;

[0015] Figure 6 is a flowchart of the related steps of the training of the gas parameter prediction model in an embodiment of the control method of the air compressor unit of the present application;

[0016] Figure 7 is a framework diagram of an embodiment of the air compressor control system of the present application;

[0017] Figure 8 is a framework diagram of an embodiment of the electronic device of the present application;

[0018] Figure 9 is a framework diagram of an embodiment of the computer readable storage medium of the present application. Detailed Implementation

[0019] To make the purpose, technical solution and effects of this application clearer and more explicit, the following describes this application in further detail with reference to the accompanying drawings and embodiments.

[0020] In this article, the term "and / or" simply describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship. Furthermore, "more" in this article means two or more objects.

[0021] It is understood that the methods of this application may include the methods provided by any of the following method embodiments and any combination of the following method embodiments that do not conflict.

[0022] It is understood that the control method of the air compressor unit in this application can be executed by any electronic device with processing capabilities, such as mobile phones, computers, tablets, etc.

[0023] Please see Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the control method for an air compressor unit according to this application. The method includes:

[0024] Step S110: Obtain the gas characteristic parameters of the working environment of the air compressor unit.

[0025] It should be noted that the air compressor unit in this application can be used to supply air to the production site; that is, the operating environment of the air compressor unit can be the production site, with the air compressor unit acting as the air production end and the operating environment as the air consumption end. Gas characteristic parameters are parameters used to describe the characteristics of the gas, and can include various parameters, such as air pressure and gas flow rate. The gas characteristics of the operating environment of the air compressor unit may also change over time; therefore, the gas characteristic parameters include the first gas parameter of the operating environment at the current moment and the second gas parameter at the next moment.

[0026] For example, the first gas parameter may include at least one of the current gas pressure and the current gas flow rate, and the second gas parameter may include at least one of the predicted gas pressure and the predicted gas flow rate at the next time step.

[0027] Step S120: At least one of the first gas parameter and the second gas parameter is detected to be outside the preset range.

[0028] The preset range can be set according to actual needs, and the preset range can be the fluctuation range of gas parameters in an acceptable working environment.

[0029] It can be understood that the step S120 can specifically be detecting that at least one of the first gas parameter and the second gas parameter is out of the preset range. Specifically, it can be that the first gas parameter is detected alone, and it is detected that the first gas parameter is out of the preset range; it can also be that the second gas parameter is detected alone, and it is detected that the second gas parameter is out of the preset range; it can also be that the first gas parameter and the second gas parameter are detected, and if any of the two is out of the preset range, it can be considered that the peculiar characteristic parameter is out of the preset range. Further specifically, the first gas parameter can be detected first, if it is out of the preset range, the second gas parameter is not detected, and if the first gas parameter is not out of the preset range, the second gas parameter is detected.

[0030] It should be noted that the first gas parameter and the second gas parameter in the step S120 can be the same type of gas parameter, for example, the first gas parameter and the second gas parameter are both air pressure or both gas flow. In addition, the second gas parameter used in the step S120 and the second gas parameter used in the step S130 can be the same type of gas parameter or different types of gas parameters, for example, the second gas parameter used in the step S120 is the predicted air pressure, and the second gas parameter used in the step S130 is the predicted gas flow.

[0031] Step S130: obtaining the state control information of the target sub-air compressor unit by using the second gas parameter of the working environment.

[0032] It should be noted that the second gas parameter of the working environment can be used to calculate the expected gas flow of the working environment at the next moment, and the state control information of the target sub-air compressor unit can be deduced from the expected gas flow. The state control information can be used to determine the use state of each air compressor in the target sub-air compressor unit at the next moment.

[0033] Specifically, the target sub-air compressor unit includes at least one of a variable frequency air compressor and a fixed frequency air compressor. For the variable frequency air compressor, the state control information can be its working frequency. For the fixed frequency air compressor, the state control information can be its use state as loading or unloading.

[0034] Among them, the target sub-air compressor unit includes at least part of the air compressors in the air compressor unit, and the state control information is used to make the gas flow at the next moment reach the target gas flow.

[0035] Step S140: controlling the use state of each air compressor in the target sub-air compressor unit based on the state control information.

[0036] It should be noted that the device in the present application can be polled once every interval of a preset time, and the related steps of the control method of the air compressor unit are executed. The purpose of executing the above steps at the current time is to determine the state control information of the air compressor unit, and the state control information is used to control the air compressor unit, so that the gas parameters of the action environment at the next time are kept within the preset range.

[0037] If the first gas parameter and the second gas parameter are detected to be out of the preset range when the related steps of the control method of the air compressor unit are executed at the current time, the use state of the air compressor unit does not need to be adjusted, and the control is ended, and the next control is waited for at the next time.

[0038] In the above scheme, when at least one of the first gas parameter and the second gas parameter of the action environment of the air compressor unit is detected to be out of the preset range, the state control information of the target sub-air compressor unit is obtained by using the second gas parameter of the action environment at the next time. Based on the state control information, the use state of each air compressor in the target sub-air compressor unit is controlled, so that the gas flow of the target sub-air compressor unit at the next time reaches the target gas flow. By the above-mentioned manner, the second gas parameter of the action environment at the next time is predicted and used to control the air compressor unit. Compared with the control using the first gas parameter, the control is performed early, and the gas parameter is still increased according to the original trend in a short time, so that the gas parameter is out of limit. The gas parameter is accurately and timely controlled, the gas parameter of the action environment is controlled in a more accurate range, and the accuracy and timeliness of the air compressor unit control are improved.

[0039] Please refer to Figure 2 , Figure 2 is a flowchart of another embodiment of the control method of the air compressor unit of the present application. The method comprises:

[0040] Step S210: predicting the second gas parameter of the action environment by using the first target feature of the air compressor unit.

[0041] Step S210 is a related step of predicting the second gas parameter. It should be noted that the prediction of the gas parameter and the control of the air compressor unit can be performed independently. The prediction of the gas parameter can be performed every interval of a certain time, and the interval time can be the same as or different from the time interval of the air compressor control. For example, the prediction of the gas parameter is performed every 1s, and the control of the air compressor unit is performed every 5s. When the control of the air compressor unit is performed at the current time, the second gas parameter is predicted first and used in the control process of the air compressor unit at the current time.

[0042] The first target feature is a feature required for making a prediction, and includes at least one of a first original feature and a target extended feature. The first original feature includes at least one of an actual gas production feature of the air compression unit, a temporary fault feature, a device state feature, a planned production line scheduling feature in an action environment, a historical production feature of the production line, and a working mode feature. The target extended feature is obtained by performing a target extension operation on the first original feature. Before step S210 is performed, the first target feature needs to be obtained. If the target extended feature is included in the first target feature, the target extended feature is obtained by performing a target extension operation on at least one first original feature.

[0043] The target extension operation includes at least one of the following: performing a statistical operation on a sub-feature included in the first original feature, performing a statistical operation on a number of sub-features belonging to a preset interval in the first original feature, performing a monotonic transformation on the first original feature, performing a polynomial transformation on the first original feature, and combining at least two first original features.

[0044] It should be noted that each of the above first target features is a time sequence feature value sequence, that is, one first target feature includes a plurality of sub-features corresponding to a plurality of time points, and each sub-feature reflects feature information of a corresponding time point. For example, the actual gas production feature includes actual gas production amounts corresponding to a plurality of reference time points, wherein the plurality of reference time points include a current time point and / or at least one historical time point; the planned production scheduling feature includes planned production amounts corresponding to different time points; the historical production feature includes planned production amounts corresponding to a plurality of reference time points; the temporary fault feature includes at least one of temporary stop line information of the production line in the action environment, material shortage information of the production line, and device fault alarm information of the air compression unit. The above three can also be embodied by corresponding information corresponding to different time points, respectively. For example, in the case of no temporary stop line of the production line, a sub-feature value corresponding to no temporary stop line information exists, and in the case of a temporary stop line of the production line, a corresponding sub-feature value is determined according to the temporary stop line situation, and is used to reflect the temporary stop line situation of the production line; the working mode feature includes working modes in different time periods, and the working mode feature can also be embodied by working mode feature values corresponding to a plurality of time points. Different working modes correspond to different working mode feature values. The working mode feature is determined based on reference information of a corresponding time period, and the reference information includes at least one of gas parameters of the action environment, production scheduling parameters, and production information of the production line.

[0045] For the target extended feature, the target extended feature includes a plurality of target extended sub-features, and the time granularity of the target extended feature is consistent with the first original feature, that is, the target extended feature / first original feature includes a plurality of sub-features / target extended sub-features corresponding to the same time point. Taking the statistics of the sub-features contained in the first original feature as an example, the sub-features contained in the first original feature are grouped, the statistics of each group are obtained, and the statistics of each group are taken as the target extended sub-features corresponding to each time point in the group.

[0046] Specifically, the first target feature includes all features required for predicting a gas parameter, such as gas flow and gas pressure, and all the first target features can also correspond to a label, which is used to determine whether the first target feature is used for predicting a gas parameter, for example, whether the first target feature is used for predicting gas flow or for predicting gas pressure. Some first target features can be used in the prediction process of multiple different types of gas parameters. For example, if the gas parameter to be predicted is gas pressure, the actual gas production feature can be an actual gas pressure feature, and if the gas parameter to be predicted is gas flow, the actual gas production feature can be an actual gas flow feature.

[0047] Step S210 can be implemented by a gas parameter prediction model, for example, a multivariate time series prediction model. The first target feature of the air compressor unit is input into the multivariate time series prediction model, and the second gas parameter of the working environment can be obtained. The second gas parameter can include a predicted gas pressure or a predicted gas flow.

[0048] Specifically, the multivariate time series prediction model can be an LSTM model (long short-term memory).

[0049] Step S220: Obtain the gas feature parameter of the working environment of the air compressor unit.

[0050] It should be noted that the obtaining of the gas characteristic parameter can be further divided into obtaining a first gas parameter and obtaining a second gas parameter, the first gas parameter is measured, and the second gas parameter is predicted. The step S230 can also be further divided into detecting whether the first gas parameter exceeds the preset range and detecting whether the second gas parameter exceeds the preset range. In some embodiments, the first gas parameter and the second gas parameter can be obtained in step S220 and used in subsequent detection of whether the range is exceeded and prediction. In some embodiments, one of the first gas parameter and the second gas parameter can be obtained first, and whether it exceeds the preset range is detected, and then the other one is obtained, and whether it exceeds the preset range is detected. In some embodiments, if only one of the first gas parameter and the second gas parameter is detected in step S230, only the one that needs to be detected can be obtained in step S220. If step S240 needs to be performed, the second gas parameter can also be obtained after step S230.

[0051] Step S230: At least one of the first gas parameter and the second gas parameter exceeds the preset range.

[0052] The gas characteristic parameter includes a first gas parameter of the working environment at the current time and a second gas parameter at the next time. If both are included, first, whether the first gas parameter exceeds the preset range is judged. If the first gas parameter exceeds the preset range, the judgment of the second gas parameter is not performed, and step S240 is directly executed. If the first gas parameter does not exceed the preset range, whether the second gas parameter exceeds the preset range is judged. If the second gas parameter exceeds the preset range, step S240 is executed. If the second gas parameter does not exceed the preset range, the control is ended. In this embodiment, the gas characteristic parameter in step S230 is taken as the air pressure as an example. The first gas parameter is the actual air pressure at the current time, and the second gas parameter is the predicted air pressure at the next time. First, the actual air pressure at the current time is obtained, and whether the actual air pressure exceeds the preset range is judged. If it exceeds, the predicted air pressure at the next time is not judged, and step S240 is directly executed. If it does not exceed, whether the predicted air pressure at the next time exceeds the preset range is judged. If it exceeds, step S240 is executed. If it does not exceed, the control is ended.

[0053] Step S240: Using the second gas parameter of the working environment and the target gas parameter, the expected gas production amount at the next time is obtained.

[0054] It should be noted that the target gas parameter can be a target gas flow, that is, a gas flow expected to be reached at the next moment, and the target gas flow is a gas flow that can make the second gas parameter of the working environment at the next moment reach the preset range. The gas production amount refers to the amount of gas produced by the air compressor unit, and the expected gas production amount refers to the gas production amount controlled by the air compressor unit in order to make the second gas parameter of the working environment at the next moment reach the preset range. In this embodiment, the second gas parameter in step S240 is taken as an example to illustrate the predicted gas flow.

[0055] In some embodiments, the above prediction can also not be performed, and a pre-upper limit of the gas parameter of the working environment is set, which further narrows the range on the basis of the preset range. When the gas parameter of the working environment exceeds the pre-upper limit without exceeding the preset range, the slope of the gas parameter fluctuation at the previous several moments is calculated. If the slope is greater than a predetermined slope, it can be considered that the gas parameter may continue to change and exceed the preset range in the next time, and then the subsequent control step is executed to control the air compressor unit. Wherein, the gas parameter can be pressure or flow, the air compressor unit and the plant equipment of the working environment form a gas production and consumption closed loop, and the pressure and flow fluctuation has certain correlation.

[0056] Step S250: based on the expected gas production amount, determine the state adjustment information of at least one air compressor in the target sub-air compressor unit as the state control information.

[0057] Wherein, the target sub-air compressor unit is the air compressor unit included in the control range in this control, which includes at least one air compressor available for control, so as to be able to adjust the gas production amount. The expected gas production amount is the gas amount expected to be produced by the target sub-air compressor unit, and according to the expected gas production amount, it can be deduced that the air compressor of the target sub-air compressor unit should be in what state, so as to be able to produce the expected gas production amount, and the state adjustment information is obtained.

[0058] Step S260: based on the state control information, control the use state of each air compressor in the target sub-air compressor unit.

[0059] Through step S260, the target sub-air compressor unit can then produce the expected gas production amount, so as to make the gas flow of the working environment reach the target gas flow, so as to realize the control of the second gas parameter at the next moment within the preset range, and complete this control.

[0060] Please refer to Figure 2 and Figure 3 , Figure 3 is a flowchart of another embodiment of step S250 of the present application, and step S250 includes:

[0061] Step S351: judging whether the preset type of air compressor is adjusted alone to make the gas production amount of the next moment reach the expected gas production amount.

[0062] It should be noted that the preset type of air compressor can be a variable frequency air compressor, the target sub-air compressor set can include several variable frequency air compressors, adjusting the frequency of the variable frequency air compressor can change the gas production amount of the air compressor, in order to simplify the control of the air compressor set and avoid too many loading and unloading steps for the air compressor, the variable frequency air compressor can be in a constant loading state, and adjusting the frequency can adjust the gas production amount. First, it is judged whether the variable frequency air compressor in the target sub-air compressor set can meet the adjustment requirement, that is, whether the gas production amount of the next moment can reach the expected gas production amount after the variable frequency air compressor is adjusted alone.

[0063] If the judgment result is yes, step S352 is executed, and if the judgment result is no, step S353 is executed.

[0064] Please refer to Figure 2 , Figure 3 and Figure 4 , Figure 4 is a flowchart of another embodiment of step S351 of the present application, and step S351 includes:

[0065] Step S4511: taking the difference between the expected gas production amount of the next moment and the actual gas production amount of the current moment as the target adjustment amount.

[0066] The difference between the expected gas production amount of the next moment and the actual gas production amount of the current moment is the gas change amount to be obtained by adjusting the air compressor, so that the difference is taken as the target adjustment amount, and the maximum gas production change value, that is, the maximum gas production change value, brought by adjusting the variable frequency air compressor is compared with the target adjustment amount, so that it can be determined whether the variable frequency air compressor can make the gas production amount of the next moment reach the expected production amount.

[0067] Step S5412: judging whether the maximum gas production change value of the preset type of air compressor is greater than or equal to the target adjustment amount.

[0068] It should be noted that whether the preset type of air compressor can meet the demand is determined by comparing the maximum value of the gas production change value obtained by adjusting the preset type of air compressor with the target adjustment amount. Specifically, the expected gas production amount of the next moment can be greater than or less than the actual gas production amount of the current moment, and the frequency of the variable frequency air compressor can be adjusted to reduce the gas production amount or increase the gas production amount. For the variable frequency air compressor, it has a current working frequency at the current moment, so the maximum gas production change value includes the maximum gas production increase value and the maximum gas production decrease value, that is, the maximum value of the gas production increase value brought by increasing the frequency of the variable frequency air compressor and the maximum value of the gas production decrease value brought by reducing the frequency of the variable frequency air compressor.

[0069] If the expected gas production amount at the next moment is greater than the actual gas production amount at the current moment, it is determined whether the maximum gas production amount increase value is greater than or equal to the target adjustment amount, and if the expected gas production amount at the next moment is less than the actual gas production amount at the current moment, it is determined whether the maximum gas production amount decrease value is greater than or equal to the target adjustment amount.

[0070] It should be noted that, since it has been determined that at least one of the first gas parameter and the second gas parameter exceeds the preset range, the case that the expected gas production amount at the next moment is equal to the actual gas production amount at the current moment will not occur.

[0071] If the determination result is yes, step S4513 is performed, and if the determination result is no, step S4514 is performed. In this way, in the case of small fluctuations in the gas parameter, the control purpose is achieved through fine adjustment, the control of the air compressor unit is simplified, and the control efficiency is improved.

[0072] Step S4513: It is determined that the separate adjustment of the air compressors of the preset type can enable the gas production amount at the next moment to reach the expected gas production amount.

[0073] Step S4514: It is determined that the separate adjustment of the air compressors of the preset type cannot enable the gas production amount at the next moment to reach the expected gas production amount.

[0074] Step S352: Based on the expected gas production amount, state adjustment information of the air compressors of each preset type in the target sub-air compressor unit is determined as state control information.

[0075] If the separate adjustment of the air compressors of the preset type can enable the gas production amount at the next moment to reach the expected gas production amount, it is not necessary to adjust the air compressors of other types in the target sub-air compressor unit. Then, based on the expected gas production amount, the use state of each air compressor of each preset type in the target sub-air compressor unit at the next moment can be deduced. Specifically, if the separate adjustment of the variable frequency air compressor can achieve the purpose, it is not necessary to adjust other air compressors, such as the fixed frequency air compressor, and the use state of the variable frequency air compressor, i.e., the expected working frequency of each variable frequency air compressor, is the state adjustment information.

[0076] The state adjustment information can be used to perform state adjustment on the air compressors of each preset type in the target sub-air compressor unit, so that each variable frequency air compressor reaches the expected working frequency, and thus the gas production amount of the target sub-air compressor unit is the expected gas production amount, and the gas pressure of the working environment can be controlled within the preset range.

[0077] Step S353: Based on the expected gas production amount, state adjustment information of each air compressor in the target sub-air compressor unit is determined as state control information.

[0078] It should be noted that since the preset type of air compressor in the target sub-air compressor unit cannot be adjusted alone to meet the demand, the fixed frequency air compressor in the target sub-air compressor unit needs to be adjusted, or the fixed frequency air compressor is combined with the variable frequency air compressor for adjustment. Step S353 can be implemented by a mixed integer programming algorithm. Specifically, the state adjustment information of each air compressor in the target sub-air compressor unit is determined by using the mixed integer programming algorithm and the expected gas production, and is used as the state control information.

[0079] In a specific application scenario, the determination of the state control information by using the mixed integer programming algorithm specifically includes taking the total energy consumption of the target sub-air compressor unit and the working state change of each air compressor in the target sub-air compressor unit as the objective function, wherein the working state represents the on or off state. It should be noted that the variable frequency air compressor is usually in a constant on state, so the state change of each air compressor in the target sub-air compressor unit is actually the state change of each fixed frequency air compressor in the target sub-air compressor unit, and the working state of the variable frequency air compressor will not actually change. The expected gas production being less than or equal to the control gas production and the number of working state changes of the target sub-air compressor unit being less than the preset upper limit are taken as constraint conditions, and the control gas production is the target gas production of the target sub-air compressor unit in the use state corresponding to the state control information. According to the objective function and the constraint condition, the state control information of the target sub-air compressor unit is determined. The above mixed integer programming algorithm can be specifically represented by the following formula:

[0080]

[0081]

[0082]

[0083] Formula 1 is the objective function, wherein E is the total energy consumption of the air compressor unit, P i_rated is the rated power of the i-th fixed frequency air compressor, x i,t is the loading and unloading state of the i-th fixed frequency air compressor, P j_rated is the rated power of the j-th variable frequency air compressor, f j_expected is the expected working frequency of the j-th variable frequency air compressor, f j_rated is the rated frequency of the j-th variable frequency air compressor, C i is the weight corresponding to the i-th fixed frequency air compressor, S i,t is the on-off state of the i-th fixed frequency air compressor at time t, S i,t-1 is the on-off state of the i-th fixed frequency air compressor at time t-1, m is the number of fixed frequency air compressors, and n is the number of variable frequency air compressors. Formula 2 is the constraint condition, wherein F i_rated is the gas production of the i-th fixed frequency air compressor, F j_rated is the gas production of the j-th variable frequency air compressor, Fexp_pro For the desired gas production, O i,t-1 is the cumulative number of switch-on and switch-off times of the i-th fixed-frequency air compressor at time t-1, N i is the upper limit value of the cumulative number of switch-on and switch-off times of the i-th fixed-frequency air compressor. Formula 3 is the boundary, x j,t is the loading and unloading state of the j-th variable-frequency air compressor, where 0 indicates that the air compressor is in an unloading state, that is, a shutdown state, and 1 indicates that the air compressor is in a loading state, that is, a startup state, f j_min is the minimum operating frequency of the j-th variable-frequency air compressor.

[0084] The output of the mixed integer programming algorithm is the expected loading state of each air compressor and the expected operating frequency of the variable-frequency air compressor, that is, the state control information of the target sub-air compressor group. Through the mixed integer programming algorithm, the air compressor group can be accurately controlled, so that the air compressor group consumes less energy while meeting the gas supply needs, the air compressor switches the least number of times, the efficiency of the air compressor group control is improved, and the air compressor group is protected.

[0085] It should be noted that the control method of the air compressor group of the present application can be executed by an edge server, and the edge server can be connected with an intelligent gateway through a wired or wireless network, and the intelligent gateway is used to control each air compressor in the air compressor group.

[0086] In the above scheme, when it is detected that at least one of the first gas parameter and the second gas parameter of the air compressor group exceeds the preset range, the state control information of the target sub-air compressor group is obtained by using the second gas parameter of the working environment at the next moment, and based on the state control information, the use state of each air compressor in the target sub-air compressor group is controlled to make the gas flow of the target sub-air compressor group at the next moment reach the target gas flow. When controlling, first use the preset type of air compressor for fine tuning, and then use mixed integer programming to adjust each air compressor in the target sub-air compressor group. By the above method, the second gas parameter of the working environment at the next moment is predicted and used to control the air compressor group. Compared with using the first gas parameter for control, the control is performed early, avoiding the situation that the gas parameter still increases according to the original trend in a short time and exceeds the limit, accurately and timely controlling the gas parameter, controlling the gas parameter of the working environment to a more accurate range, improving the accuracy and timeliness of the air compressor group control. When the fluctuation is not large, the air compressor of the preset type can be simply controlled alone, improving the control efficiency and avoiding excessive adjustment of the air compressor group.

[0087] Please refer to Figure 5 , Figure 5 is a flowchart of another embodiment of the control method of the air compressor group of the present application, and the method comprises:

[0088] Step S510: Obtain characteristic data of a plurality of air compressors in an air compressor set.

[0089] It should be noted that the air compressors in the air compression station are used to supply air to the production site in the present application, and the air compressor set in step S510 refers to a plurality of air compressors in the air compressor set that can be controlled. All air compressors in the air compression station may have a part of the equipment that cannot be included in the control due to the emergency stop state, fault, maintenance stage, cumulative running time exceeding the limit, or intelligent joint control switch closed, and thus is not included in the scope of the air compressor set in step S510.

[0090] Specifically, the characteristic data can include a plurality of the on-off state S, shutdown time t close , gas-electricity ratio VSP, rated power P rated , rated exhaust volume F rated , shutdown time t close and exhaust temperature T flow of the air compressor.

[0091] Step S520: Determine the state score of each of the plurality of air compressors based on the characteristic data of the plurality of air compressors.

[0092] It should be noted that the state score of each of the plurality of air compressors can be determined by the following formula, which is based on the characteristic data of the air compressor:

[0093] Score = w1*S + w2*(t close -180) + w3 / VSP + w4 / P rated + w5 / F rated + w6*t close + w7 / T flow Formula 4

[0094] wherein w i is a weight coefficient, which can be obtained from expert experience.

[0095] It should be noted that the on-off state S can be 0 or 1, 0 representing shutdown and 1 representing startup. The state score can be calculated by the above formula, wherein the characteristic data of the air compressor will affect its state score. For example, the lower the gas-electricity ratio VSP of the air compressor, the less power consumption to achieve the same load target, and the higher the state score. The closer the rated exhaust volume F rated to the production-demand difference, the higher the state score, wherein the production-demand difference can be determined according to the latest predicted production-demand difference, so as to select the air compressor to adapt to the production-demand difference and avoid overcorrection. The higher the exhaust temperature T flowThe lower the air compressor, the same gas production conditions, can reduce the temperature is too high and the load time is too long caused by equipment damage, the state score is higher. In addition, the shutdown time t close The larger the state score, the higher the rated power P rated The lower the state score, the higher.

[0096] Step S530: grouping the air compressors using the state score to obtain a to-be-controlled group and a standby control group.

[0097] Specifically, the air compressors are grouped using a preset score, which can be set according to requirements. The air compressors with a state score greater than or equal to the preset score are taken as the to-be-controlled group, and the air compressors with a state score less than the preset score are taken as the standby control group.

[0098] Step S540: taking the to-be-controlled group as a target sub-air compressor group.

[0099] It should be noted that the standby control group screened out in this way may have some equipment problems, such as oil leakage, and cannot be frequently adjusted in use. The standby control group can be used only when necessary, which can protect the air compressors and reduce the risk of equipment damage. Step S540 and step S550 are executed alternatively. Generally, step S540 is executed, and the to-be-controlled group is taken as the target sub-air compressor group. The to-be-controlled group can meet the target gas flow at the next moment. If the to-be-controlled group cannot meet the target gas flow at the next moment, step S550 is executed, and the to-be-controlled group and the standby control group are taken as the target sub-air compressor group.

[0100] It should be noted that for air supply using the target sub-air compressor group, the air compressors in the target sub-air compressor group can also be prioritized according to the above state score. The prioritization result can also be used to assist in controlling the target sub-air compressor group and to determine which air compressor to preferentially select.

[0101] Step S550: taking the to-be-controlled group and the standby control group as the target sub-air compressor group.

[0102] Step S560: obtaining a gas characteristic parameter of an operating environment of the air compressor group.

[0103] Step S570: detecting that at least one of the first gas parameter and the second gas parameter exceeds a preset range.

[0104] Step S580: obtaining state control information of the target sub-air compressor group using the second gas parameter of the operating environment.

[0105] Step S590: controlling the use state of each air compressor in the target sub-air compressor group based on the state control information.

[0106] The related descriptions of steps S560-S590 can refer to the foregoing related content about steps S110-S140, and will not be described here.

[0107] It should be noted that the control method of the air compressor unit can be executed by an edge server, which can control the intelligent gateway of each air compressor in the air compressor unit. The edge server is deployed with an air compressor unit joint control main program and algorithms such as air pressure, flow prediction, and mixed integer programming, so as to execute the control method of the air compressor unit and control the air compressor unit. The edge server can be connected with a cloud server, and the cloud server is deployed with an air compressor unit energy-saving intelligent control business operation system, which can be used by users to perform air compressor control operations and monitor and display the control situation of the air compressor unit. The cloud server can determine the uncontrolled air compressors in the air compressor station and thus determine the air compressor unit in response to user operations, so that the edge server can determine the air compressor unit for subsequent control operations.

[0108] In the above scheme, the air compressors are screened to select the to-be-controlled group and the alternative control group, and the target sub-air compressor unit is selected according to needs. When at least one of the first gas parameter and the second gas parameter of the working environment of the air compressor unit is detected to be out of the preset range, the second gas parameter of the working environment at the next moment is used to obtain the state control information of the target sub-air compressor unit. Based on the state control information, the use state of each air compressor in the target sub-air compressor unit is controlled, so that the gas flow of the target sub-air compressor unit at the next moment reaches the target gas flow. Through the above manner, suitable air compressors can be selected as control objects, unnecessary use state changes are avoided, and the air compressors are protected. The second gas parameter of the working environment at the next moment obtained by prediction is used to control the air compressor unit. Compared with using the first gas parameter for control, the control is performed early, the gas parameter is prevented from exceeding the limit due to the original trend of the gas parameter increasing in a short time, the gas parameter is accurately and timely controlled, and the accuracy and timeliness of the air compressor unit control are improved.

[0109] Please refer to Figure 6 , Figure 6 is a flowchart of the related steps of the gas parameter prediction model training in an embodiment of the control method of the air compressor unit of the present application. The related steps of the gas parameter prediction model training need to be completed first before the gas parameter prediction of the next moment of the air compressor unit is performed. The related steps of the gas parameter prediction model training can include:

[0110] Step S610: Obtain a plurality of second original features.

[0111] The second original feature and the extended feature involved below are features affecting the air pressure and flow of the air compressor unit, and the second original feature refers to a feature without extension processing. Specifically, the second original feature can include at least one of an actual gas production feature of the air compressor unit, a temporary fault feature, a device state feature, a planned production feature of a production line in an action environment, a historical production feature of the production line, and a working mode feature.

[0112] The features in the present application each include a series of sub-features determined according to a uniform time interval for each interval as a sub-feature corresponding to each time. For example, the actual gas production feature of the air compressor unit can include the air pressure P c The air pressure P c as a sub-feature, thereby composing the feature. The time interval (time granularity) here can be determined according to user needs, for example, 1s, 10s, etc. The time interval can be consistent with the time interval of each air compressor unit control polling or can not be consistent. The time interval is also the time interval for gas parameter prediction.

[0113] Specifically, the FEMS (Factory Energy Management System) system collects feature information from the APS (Advanced Planning and Scheduling) system, the MES (Manufacturing Execution System) system, the EAM (Enterprise Asset Management) system, and the SCADA (Supervisory Control And Data Acquisition) system, etc. and processes the feature information to obtain the second original feature. For example, the device collected feature information is time series data, and the shift function in pandas is used to design a method to convert the time series data into supervised data. First, the time series data is resampled according to a certain time granularity, that is, sampling at a preset time interval. Under the unified time axis, the supervised data length n_in is selected, and the resampled time series data is processed before and after the shift to convert it into supervised data. The supervised data is the feature described in the present application, and each data included in the supervised data is a sub-feature.

[0114] Further, the actual gas production feature of the air compressor unit can include at least one of the air pressure P c the flow F c and other features for representing actual gas production. The actual gas production feature can include actual gas production features corresponding to a plurality of times, respectively.

[0115] The temporary fault feature can include at least one of a line temporary stop information Is_StopLine in the working environment, a procurement part work order complete set material shortage detail situation Pur_Material, an equipment fault alarm Alram information (i.e., air compressor fault), a fault expected repair time Repair_Time, and the like, which are used to represent some temporary faults that can affect the next time air production of the air compressor unit. For example, if the line temporary stop occurs, the gas consumption in the working environment decreases, and the next time air production of the air compressor unit is reduced. The procurement part work order complete set material shortage detail situation Pur_Material, the equipment fault alarm Alram information, and the fault expected repair time Repair_Time all affect the gas consumption in the working environment, thereby affecting the next time air production of the air compressor unit. The temporary fault feature can include a plurality of temporary fault features corresponding to a plurality of times, respectively. Specifically, the fault feature value can be used to represent the temporary fault situation at the current time.

[0116] The equipment state feature can include at least one of an on-off state OC_Status, a health state Alarm_Status, a loading time O_Time, an oil pressure Oil_Time, a running time Run_Time, an exhaust temperature R_Temperatue, and the like, which are used to represent the features of the air compressor unit. The equipment state feature can be obtained based on the state features of a plurality of air compressors. The equipment state feature includes a plurality of equipment state features corresponding to a plurality of times, respectively. The equipment state feature value can represent the equipment state situation at the current time.

[0117] The planned production feature of the line in the working environment can include at least one of a planned production amount Product_O_All, which is used to represent the production plan in the working environment. Specifically, the inventory organization ID Inv, the workshop ID Gid, the work order code Mcode, and the planned production amount Product_O, and the like, which are used to represent the production plan in the working environment, are obtained from the APS system. For example, the production plan information of different factories in different time periods is obtained. The planned production amount Product_O_All of the working environment is obtained based on the above production plan information. Then, resampling is performed. The planned production amount obtained after resampling includes a plurality of planned production amounts Product_O_All corresponding to a plurality of times, respectively.

[0118] The historical output characteristics of the production line may include at least one of the characteristics used to characterize the gas consumption in the operating environment, such as the historical gas production volume Product_H_All of the operating environment. Specifically, the historical output Product_H corresponding to different plants is obtained from the MES system, and integrated into the overall historical gas production volume Product_H_All of the operating environment based on the plant scope of the operating environment. Then, resampling is performed to obtain the historical gas production volume Product_H_All corresponding to several time points.

[0119] The working model features include the working modes at different times, which can also be reflected by the working mode feature values ​​corresponding to several times. Different working modes correspond to different working mode feature values. The working mode features are determined based on reference information for the corresponding time period. The reference information includes at least one of the following: gas parameters of the working environment, production scheduling parameters, and production line output information. Specifically, clustering algorithms can be used to process the reference information to obtain the working modes corresponding to different times. The clustering algorithm can be an artificial intelligence algorithm, such as the K-Means algorithm, DBSCAN algorithm, hierarchical clustering method, etc. The gas pressure at the gas-using end, instantaneous flow rate, relevant production scheduling parameters in the APS system, and real-time output information of MES are used as inputs to the clustering algorithm. The working modes can be divided into holidays (production stoppage), holidays (light production), normal production, and daily lunch break / afternoon tea (light production). The same working mode corresponds to the same working mode features.

[0120] Step S620: Perform a preset extension on at least one second original feature to obtain an extended feature.

[0121] Preset expansion may include at least one of the following: counting the sub-features contained in the second original feature, counting the number of sub-features in the second original feature that belong to a preset interval, performing a monotonic transformation on the second original feature, performing a polynomial transformation on the second original feature, or combining at least two second original features.

[0122] It should be noted that the final obtained extended features include several extended sub-features. The temporal granularity of the extended features is consistent with that of the second original features before extension. In other words, the extended features / second original features include the same sub-features / extended sub-features corresponding to several time points. One extended sub-feature corresponds to one time point, and the time interval between time points is consistent with the unified time interval corresponding to the second original features. The number of sub-features of the extended features is also consistent with the number of sub-features of the second original features.

[0123] Further, the statistics of the sub-features contained in the second original features can include grouping the sub-features contained in the second original features, and performing statistics on each group of sub-features, and the statistics can include at least one of the mean, median, mode, maximum value, minimum value, and standard deviation. One type of statistics obtained by performing statistics can be used as an extended feature. Statistics are performed on each group, and the statistics of each group are used as the corresponding extended sub-features of each time in the group, that is, the statistics of each group are used to replace all sub-features in the group, and after all replacements are completed, the extended features are obtained. For example, the pressure P c For example, if the feature includes P1-P4, the pressure of the working environment corresponding to two time points is used as a group, the average value of each group is obtained, that is, and The extended sub-features corresponding to two time points in each group are the average value of the group, and the final extended features are

[0124] The statistics of the number of sub-features belonging to the preset interval in the second original features can include grouping the sub-features contained in the second original features, and then counting the number of sub-features belonging to the preset interval in each group of sub-features, that is, frequency statistics. Then, the statistics of each group are used as the corresponding extended sub-features of each time in the group, that is, the statistics of each group are used to replace all sub-features in the group, and after all replacements are completed, the extended features are obtained. For example, the second original features are a series of sub-features, and all sub-features are grouped by hour, and the number of times that the sub-features exceed 80 per hour is counted.

[0125] Step S630: filtering the second original features and the extended features based on the importance to obtain the second target features.

[0126] Step S630 can be implemented by a gradient boosting decision tree. Specifically, the gradient boosting decision tree can be used to filter features with an importance greater than a preset value from the second original features and the extended features, and the features are used as the second target features.

[0127] It should be noted that the second target features are used to train the prediction model, so that the trained prediction model can be used to predict the gas parameters using the target features. When using the model, the first target features, that is, the target feature values required for the current prediction, need to be obtained.

[0128] The second target feature can include part of the second original feature / extended feature, and the category of the first original feature in the first target feature corresponds to the category of the second original feature in the second target feature. The second target feature can include part of all second original feature categories. The extended feature included in the second target feature can not be obtained by extending the second original feature in the second target feature. For example, the second original feature includes A and B, C and D are obtained by extending A, and the second target feature includes B and C.

[0129] In a specific application scenario, the GBDT model is used for feature selection. Specifically, the GBDT model can be used to determine the importance of each feature. The average importance of feature j in multiple trees is:

[0130]

[0131] where M is the total number of decision trees in the GBDT model, T m represents the mth decision tree, represents the importance of feature j output by the mth decision tree.

[0132] The average importance of each feature in multiple trees is compared with a preset value, and the features with an average importance greater than the preset value are selected as second target features.

[0133] Step S640: Obtain at least part of the sub-feature from each second target feature to obtain the sample feature corresponding to each second target feature.

[0134] It should be noted that in the gas parameter prediction model training stage, all features include a series of sub-features corresponding to each time in a period of time. The device can select a certain time as the current time, and the sub-features before that time are historical sub-features. After determining the current time, the device can select the sub-features of the current time and a plurality of historical sub-features from each second target feature as sample features corresponding to each second target feature. For example, the second target feature includes A and B, A includes A1-A 10 , and B includes B1-B 10, determine the current time and the second target feature corresponding to the current time as A5 and B5, at this time the prediction model needs to predict the air pressure and flow at the next time, that is, the air pressure and flow at t6, the number of sub-features used by the prediction model is certain, the sample features of the current time are selected according to the number of sub-features used by the prediction model, for example, the prediction model needs to use 4 sub-features, then A2-A5 and B2-B5 are selected from A and B as the sample features corresponding to the second target feature. At this time, the gas parameter prediction model obtained by training is also used to predict the gas parameters at the next time by using 4 sub-features. The number of features and the number of sub-features used by the model for prediction are only for illustration, and the actual number can be determined by the user's needs.

[0135] After the above operation for many times, a plurality of sample features can be obtained, that is, a plurality of time points are selected as the current time, and the sample features are determined respectively, so that a plurality of sample features are obtained, which are used as samples for training the gas parameter prediction model.

[0136] Step S650: training the gas parameter prediction model using a plurality of sample features.

[0137] The gas production feature corresponding to the next time of the current time is the standard prediction value expected to be output by the gas parameter prediction model. A sample feature is input into the gas parameter prediction model, and a sample prediction result is output. The sample prediction result is compared with the standard prediction value, the parameters of the gas parameter prediction model are adjusted, and the obtained gas parameter prediction model after training can be used in the control method of the air compressor unit.

[0138] It should be noted that the training of the gas parameter prediction model is performed using the second target feature obtained after importance screening, so the target feature required by the gas parameter prediction model for gas prediction has been determined, and the second target feature is used for training during training, so the target feature required by the prediction model can be determined. In the process of use, the first target feature is used for prediction. It can be understood that the first target feature and the second target feature are different values corresponding to the target feature required by the prediction model. For example, the actual gas production feature obtained after screening is used as the second target feature required by the model for prediction, and the first target feature used when using the model for prediction also includes the actual gas production feature obtained after expansion. The values of the actual gas production features in the first target feature and the second target feature are different. The second target feature can include second original features and / or expansion features. The expansion features in the second target feature also correspond to the target expansion features, which are all based on the same type of original features after target expansion. After the gas parameter prediction model is trained, the features required by the prediction model are determined. When using the model, only the first target feature corresponding to the second target feature needs to be obtained. Specifically, if the second target feature includes some second original features, the first original features corresponding to these second original features need to be obtained. If the second target feature includes some expansion features, which are obtained by target expansion of some second original features, only the corresponding first original features need to be target expanded to obtain the target expansion features.

[0139] Through the above steps, a plurality of second original features are obtained; at least one second original feature is expanded by a preset expansion to obtain an expansion feature; the plurality of second original features and the expansion feature are screened based on an importance to obtain a second target feature; a plurality of sample features are generated based on the second target feature; and the gas parameter prediction model is trained based on the plurality of sample features. The production planning, temporary fault information and the like can be integrated into the features used for prediction, and the features are expanded and screened, so that the prediction model finally trained can accurately predict the change of the gas parameter.

[0140] Please refer to Figure 7 , Figure 7 is a schematic diagram of an embodiment of the air compressor control system of the present application.

[0141] In this embodiment, the air compressor control system 70 includes an air compressor set 71 and a control device 72, wherein the air compressor set 71 can include several air compressors, which can include variable frequency air compressors and fixed frequency air compressors. The control device 72 can control the state of each air compressor in the air compressor set 71, and the control device 72 can execute any of the above-mentioned air compressor set control methods to control the use state of at least one air compressor in the air compressor set 71.

[0142] Please refer to Figure 8 , Figure 8 is a schematic diagram of the framework of an embodiment of the electronic device of the present application.

[0143] In this embodiment, the electronic device 80 includes a memory 81 and a processor 82, wherein the memory 81 is coupled to the processor 82. Specifically, each component of the electronic device 80 can be coupled together through a bus, or the processor 82 of the electronic device 80 is connected to each of the other components one by one. The electronic device 80 can be any device with processing capability, such as a computer, a tablet computer, a mobile phone, etc., and the electronic device 80 can serve as the aforementioned control device 72.

[0144] The memory 81 is used to store program data executed by the processor 82 and data in the processing process of the processor 82, etc. For example, GBDT model, etc. The memory 81 includes a non-volatile storage portion for storing the above-mentioned program data.

[0145] The processor 82 controls the operation of the electronic device 80, and the processor 82 can also be referred to as a CPU (Central Processing Unit). The processor 82 can be an integrated circuit chip with signal processing capability. The processor 82 can also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. In addition, the processor 82 can be implemented by multiple integrated circuit chips together.

[0146] The processor 82 calls the program data stored in the memory 81 to execute instructions to implement any of the above-mentioned air compressor set control methods.

[0147] Please refer to Figure 9 , Figure 9 is a schematic diagram of the framework of an embodiment of the computer readable storage medium of the present application.

[0148] In this embodiment, the computer readable storage medium 90 stores program data 91 executable by the processor, which can be executed to implement any of the above-mentioned air compressor set control methods.

[0149] The computer readable storage medium 90 can specifically be a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc. which can store program data, or can also be a server which stores the program data and sends the stored program data to other devices for running, or can also run the stored program data by itself.

[0150] In some embodiments, the computer readable storage medium 90 can also be a memory as shown in Figure 8

[0151] The above description is merely an embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made by using the content of the present application specification and drawings, or directly or indirectly applied to other related technical fields, are also included in the patent protection scope of the present application.​

Claims

1. A control method for an air compressor unit, characterized in that, The method includes: The gas characteristic parameters of the working environment of the air compressor unit are obtained, wherein the gas characteristic parameters include a first gas parameter of the working environment at the current time and a second gas parameter at the next time, and the second gas parameter includes a predicted gas pressure or a predicted gas flow rate. At least one of the first gas parameter and the second gas parameter is detected to be outside a preset range; Using the second gas parameter, the state control information of the target sub-air compressor unit is obtained, wherein the target sub-air compressor unit includes at least one air compressor in the air compressor unit; obtaining the state control information of the target sub-air compressor unit using the second gas parameter includes: using the second gas parameter and the target gas parameter to obtain the expected gas production at the next moment, wherein the target gas parameter is the gas parameter expected to be achieved at the next moment; based on the expected gas production, determining the state adjustment information of at least one air compressor in the target sub-air compressor unit as the state control information; Based on the status control information, the operating status of each air compressor in the target sub-air compressor unit is controlled.

2. The method according to claim 1, characterized in that, Obtaining the second gas parameters of the operating environment of the air compressor unit includes: Using the first target characteristics of the air compressor unit, the second gas parameters of the operating environment are predicted.

3. The method according to claim 2, characterized in that, The first target feature includes at least one of the first original feature and the target extended feature; The first original feature includes at least one of the following: the actual gas production feature of the air compressor unit, the temporary fault feature, the equipment status feature, the planned production scheduling feature of the production line in the working environment, the historical output feature of the production line, and the working mode feature.

4. The method according to claim 2, characterized in that, The step of predicting the second gas parameters of the operating environment using the first target characteristics of the air compressor unit includes: The first target characteristics of the air compressor unit are predicted using a gas parameter prediction model to obtain the second gas parameters of the operating environment, wherein the gas parameter prediction model is a multivariate time series prediction model.

5. The method according to claim 3, characterized in that, Before predicting the second gas parameters of the operating environment using the first target characteristics of the air compressor unit, the method further includes: At least one of the first original features is expanded to obtain the target expanded features; The first original feature and the target extended feature are used as the first target feature.

6. The method according to claim 5, characterized in that, The target expansion of at least one of the first original features includes at least one of the following: Statistical analysis is performed on the sub-features contained in the first original feature; The number of sub-features belonging to the preset interval in the first original feature is counted; Perform a monotonic transformation on the first original feature; Perform a polynomial transformation on the first original feature; Combine at least two of the first original features.

7. The method according to claim 1, characterized in that, The step of determining the status adjustment information of at least one air compressor in the target sub-air compressor unit based on the expected gas production includes: Determine whether adjusting the preset type of air compressor alone can make the air output at the next moment reach the desired air output; If possible, then based on the expected gas production, determine the status adjustment information of each preset type of air compressor in the target sub-air compressor unit; If not, then based on the expected gas production, determine the status adjustment information of each air compressor in the target sub-air compressor unit.

8. The method according to claim 7, characterized in that, The determination of whether individually adjusting a preset type of air compressor can make the air output at the next moment reach the desired air output includes: The difference between the expected gas production at the next moment and the actual gas production at the current moment is used as the target adjustment amount; Determine whether the maximum air output change value of the preset type of air compressor is greater than or equal to the target adjustment value; If so, then it is determined that adjusting the preset type of air compressor individually can make the gas production at the next moment reach the desired gas production. If not, adjusting the preset type of air compressor alone cannot make the air output at the next moment reach the desired air output.

9. The method according to claim 7, characterized in that, The determination of the state adjustment information of each air compressor in the target sub-air compressor unit based on the expected gas production includes: The total energy consumption of the target sub-air compressor unit and the changes in the working status of each air compressor in the target sub-air compressor unit are used as the objective function, where the working status represents the on or off state. The expected gas production is less than or equal to the controlled gas production, and the number of times the target sub-air compressor unit changes its operating state is less than a preset upper limit are used as constraints. The controlled gas production is the gas production of the target sub-air compressor unit in the operating state corresponding to the state control information. Based on the objective function and the constraints, the state adjustment information of each air compressor in the target sub-air compressor unit is determined.

10. The method according to claim 1, characterized in that, The method further includes: Obtain characteristic data of several air compressors in the air compressor unit; Using the characteristic data of the aforementioned air compressors, the state score of each air compressor is determined. The air compressors are grouped using the state scores to obtain a control group and a candidate control group. If the control group can meet the target gas flow rate at the next moment, the control group will be designated as the target sub-air compressor group. If the control group cannot meet the target gas flow rate at the next moment, the control group and the alternative control group will be used together as the target sub-compressor unit.

11. The method according to claim 4, characterized in that, Before predicting the first target characteristics of the air compressor unit using a gas parameter prediction model to obtain the second gas parameters of the operating environment, the method further includes: Obtain several second primitive features; At least one of the second original features is extended in a predetermined manner to obtain an extended feature; The second target feature is obtained by filtering the aforementioned second original features and extended features based on their importance. At least some sub-features are obtained from each of the second target features to obtain the sample features corresponding to each of the second target features; The gas parameter prediction model is trained using several of the sample features.

12. The method according to claim 11, characterized in that, The second target features obtained by filtering the several second original features and extended features based on importance include: Using a gradient boosting decision tree, features with importance greater than a preset value are selected from the several second original features and extended features, and these are used as the second target features.

13. The method according to claim 11, characterized in that, The aforementioned second primary characteristics include at least one of the following: actual air production characteristics of the air compressor unit, temporary fault characteristics, equipment status characteristics, planned production scheduling characteristics of the production line in the operating environment, historical output characteristics of the production line, and operating mode characteristics.

14. The method according to claim 11, characterized in that, The pre-defined extension of at least one of the second original features includes at least one of the following: Statistical analysis is performed on the sub-features contained in the second original feature; The number of sub-features belonging to the preset interval in the second original feature is counted; Perform a monotonic transformation on the second original feature; Perform a polynomial transformation on the second original feature; Combine at least two of the second original features.

15. An air compressor control system, characterized in that, The air compressor control system includes an air compressor unit and a control device, wherein the control device performs the method as described in any one of claims 1-14 to control the operating status of at least one air compressor in the air compressor unit.

16. An electronic device, characterized in that, The electronic device includes a processor and a memory, the memory being used to store program data, and the processor being used to execute the program data to implement the method as described in any one of claims 1-14.

17. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store program data that can be executed to implement the method as described in any one of claims 1-14.

Images