Air compressor adjusting and optimizing method and system based on artificial intelligence and cloud edge collaboration

Abstract

The invention provides an air compressor adjusting and optimizing method and system based on artificial intelligence and cloud edge collaboration. The method comprises the steps that a preset full-connection neural network algorithm is applied to conduct air compressor performance model modeling, a power-yield neural network model and a power-frequency neural network model are obtained and stored in a cloud end, and an edge end conducts frequency adjustment on an air compressor according to the models. A demand prediction model is constructed by applying a preset long-short-term memory artificial neural network algorithm and is stored in the cloud, the edge end predicts the current demand of the air compressor unit according to the demand prediction model to obtain the expected demand, the edge end calculates the expected total power of the air compressor unit when the air compressor unit reaches the expected demand, and the expected total power of the air compressor unit reaches the expected demand. And the edge end obtains the full-load flow velocity of the air compressor, compares the expected demand with the full-load flow velocity, and selects the optimal scheduling optimization scheme of the air compressor unit according to the comparison result, so that the air compressor unit can be always in the optimal efficiency state under the working condition, and the purposes of saving energy and reducing cost are achieved.

Description

Technical Field

[0001] This application relates to the technical field of intelligent optimization regulation of air compressor units. Specifically, it involves an air compressor optimization method and system based on artificial intelligence and cloud-edge collaboration Background Art

[0002] The frequency conversion control of traditional variable frequency air compressors generally uses the proportional-integral-derivative (PID) algorithm based on the set discharge pressure and the current pressure. In actual engineering, multiple air compressors generally operate in parallel. Due to problems such as different pipelines, connection positions of air storage tanks, and sensor errors, some air compressors with the same PID parameters are more sensitive, which will cause the sensitive air compressors to be overloaded and the sluggish air compressors to be underloaded. At this time, the system does not operate in an optimal state. At the same time, due to different sensitivities for different opening combinations, optimization cannot be achieved by simply adjusting the PID parameters. The existing unit scheduling generally uses the full-load method (when the power or frequency of the operating air compressors all reach the rated value, add a machine, and when the supply is greater than the demand, the air compressor automatically reduces the frequency until it unloads) or pressure control (start one when the main pipe pressure is lower than a certain threshold or close one when it is higher than a certain threshold). Such a scheduling method is blind, cannot consider the different actual efficiencies of multiple air compressors, cannot give priority to starting air compressors with high operating efficiency, cannot give priority to stopping air compressors with low operating efficiency, and cannot make the air compressor unit operate at the optimal efficiency

[0003] In view of the above problems, there is an urgent need for an effective technical solution Summary of the Invention

[0004] The present invention provides a method for frequency conversion regulation and start-stop scheduling of a variable frequency air compressor unit. The method includes using a fully connected neural network (FCNN) in the cloud to build an air compressor performance model, using a long short-term memory neural network (LSTM) to build a compressed air demand model, and performing frequency conversion optimization and scheduling optimization at the edge according to the performance model and demand model synchronized from the cloud to ensure that the air compressor unit always operates at the optimal efficiency under the working conditions, achieving the purpose of energy conservation and cost reduction

[0005] This application also provides an air compressor optimization method based on artificial intelligence and cloud-edge collaboration, including the following steps Using a preset fully connected neural network algorithm to build an air compressor performance model, obtaining a power-output neural network model and a power-frequency neural network model, and storing them in the cloud The edge side adjusts the frequency of the air compressor according to the power-output neural network model and the power-frequency neural network model Using a preset long short-term memory artificial neural network algorithm to build a demand prediction model and storing it in the cloud, and the edge side predicts the current demand of the air compressor unit according to the demand prediction model to obtain the expected demand The expected total power of the air compressor unit when the edge - side computing air compressor unit reaches the expected demand; The edge - side obtains the full - load flow rate of the air compressor, compares the expected demand with the full - load flow rate, and selects the best scheduling optimization plan for the air compressor unit according to the comparison result.

[0006] Optionally, in the air compressor optimization method based on artificial intelligence and cloud - edge collaboration described in this application, the use of the preset fully - connected neural network algorithm to model the air compressor performance model, obtain the power - output neural network model and the power - frequency neural network model, and store them in the cloud includes: Real - time collect the operation data of the air compressor, including instantaneous power, instantaneous frequency, status - marker data, and average flow rate; Based on the preset fully - connected neural network algorithm, use the instantaneous power, status - marker data, and average flow rate of a single air compressor to initialize the parameters of the preset power - output model, and obtain the initialized power - output neural network model; Calculate the total average flow rate of the air compressor unit according to the average flow rates of all air compressors; Train the initialized power - output neural network model according to the total average flow rate, the instantaneous power, and the status - marker data of each air compressor, and obtain the power - output neural network model; According to the preset fully - connected neural network algorithm and using the instantaneous power and instantaneous frequency of each air compressor for model training, obtain the power - frequency neural network model; Store the power - output neural network model and the power - frequency neural network model in the cloud.

[0007] Optionally, in the air compressor optimization method based on artificial intelligence and cloud - edge collaboration described in this application, the edge - side adjusts the frequency of the air compressor according to the power - output neural network model and the power - frequency neural network model, including: The edge - side obtains the current pressure value, current power value, current flow rate, and current status - marker data of the air compressor; Perform proportional - integral operation according to the current pressure value and the set pressure value to obtain the system power increase - decrease value, and determine the adjusted power value and the set value of the adjustment times; Input the adjusted power value and the current status - marker data into the power - output neural network model to process and obtain the adjusted flow rate; Calculate the difference between the adjusted flow rate and the current flow rate, and select the target air compressor according to the difference; Input the adjusted power value into the power - frequency neural network model to process and obtain the adjusted frequency value, set the target air compressor inverter to the adjusted frequency value, and synchronously update the current adjustment times; Compare the current adjustment times with the set value of the adjustment times. If the current adjustment times are less than the set value of the adjustment times, continue to repeat the adjustment.

[0008] Optionally, in the air compressor optimization method based on artificial intelligence and cloud-edge collaboration described in this application, the method of using the preset long short-term memory artificial neural network algorithm to construct a demand prediction model and store it in the cloud, and the edge side predicts the current demand of the air compressor unit according to the demand prediction model to obtain the expected demand, includes: Obtain the average demand of the air compressor, as well as the average flow rate and instantaneous pressure within a preset time period; Based on the preset long short-term memory artificial neural network algorithm, and according to the average flow rate, instantaneous pressure, and average demand within the preset time period, perform model training to obtain a demand prediction model and store it in the cloud; The edge side obtains the average flow rate and instantaneous pressure of the air compressor unit within the current preset time period; Input the average flow rate and instantaneous pressure within the current preset time period into the preset demand prediction model for processing to obtain the expected demand.

[0009] Optionally, in the air compressor optimization method based on artificial intelligence and cloud-edge collaboration described in this application, the edge side calculates the expected total power of the air compressor unit when it reaches the expected demand, including: The edge side obtains the efficiency corresponding to each air compressor at different powers, and selects the power value with the highest efficiency and the average flow rate as the optimal power value and the optimal output value; Sum up the optimal power values of all air compressors to obtain the optimal total power of the air compressor unit, and sum up the optimal output values of all air compressors to obtain the optimal total output of the air compressor unit; Determine the adjusted power value according to the comparison result between the expected demand and the optimal total output; Input the adjusted power value and the current status marker data into the power-output neural network model for processing to obtain the adjusted flow rate; Calculate the difference between the adjusted flow rate and the current flow rate, and select the target air compressor according to the difference; Input the adjusted power value into the power-frequency neural network model for processing to obtain the adjusted frequency value, set the target air compressor inverter to the adjusted frequency value, and calculate the adjusted average flow rate of the air compressor unit according to the adjusted flow rate of the target air compressor; Compare the adjusted average flow rate of the air compressor unit with the expected demand. If they are consistent, use the total power of the air compressor unit at this time as the expected total power of the air compressor unit; If they are inconsistent, continue to repeat the adjustment until the adjusted average flow rate of the air compressor unit reaches the expected demand, and use the total power of the air compressor unit at this time as the expected total power of the air compressor unit.

[0010] Optionally, in the air compressor optimization method based on artificial intelligence and cloud-edge collaboration described in this application, the edge device obtains the full-load flow rate of the air compressor, compares the expected demand with the full-load flow rate, and selects the best scheduling optimization plan for the air compressor unit according to the comparison result, including: If the full-load flow rate is greater than the expected demand, calculate the expected total power of the air compressor unit with one additional air compressor and two additional air compressors respectively, and select the plan with the lowest expected total power for execution; If the full-load flow rate is less than the expected demand, calculate the expected total power and energy saving rate of the air compressor unit with one additional air compressor, one less air compressor, and one replaced air compressor respectively, and select the plan with an energy saving rate greater than 5% and the highest energy saving rate for execution.

[0011] In a second aspect, this application provides an air compressor optimization system based on artificial intelligence and cloud-edge collaboration. The system includes: a memory and a processor. A program of the air compressor optimization method based on artificial intelligence and cloud-edge collaboration is stored in the memory. When the program of the air compressor optimization method based on artificial intelligence and cloud-edge collaboration is executed by the processor, the following steps are implemented: Use a preset fully connected neural network algorithm to build an air compressor performance model, obtain a power-output neural network model and a power-frequency neural network model, and store them in the cloud; The edge device adjusts the frequency of the air compressor according to the power-output neural network model and the power-frequency neural network model; Use a preset long short-term memory artificial neural network algorithm to build a demand prediction model and store it in the cloud. The edge device predicts the current demand of the air compressor unit according to the demand prediction model to obtain the expected demand; The edge device calculates the expected total power of the air compressor unit when the air compressor unit reaches the expected demand; The edge device obtains the full-load flow rate of the air compressor, compares the expected demand with the full-load flow rate, and selects the best scheduling optimization plan for the air compressor unit according to the comparison result.

[0012] Optionally, in the air compressor optimization system based on artificial intelligence and cloud-edge collaboration described in this application, the use of a preset fully connected neural network algorithm to build an air compressor performance model, obtain a power-output neural network model and a power-frequency neural network model, and store them in the cloud includes: Collect the operation data of the air compressor in real time, including instantaneous power, instantaneous frequency, status marker data, and average flow rate; Based on the preset fully connected neural network algorithm, use the instantaneous power, status marker data, and average flow rate of a single air compressor to initialize the parameters of the preset power-output model to obtain an initialized power-output neural network model; Calculate the total average flow rate of the air compressor unit according to the average flow rate of all air compressors; Train the initialized power-output neural network model according to the total average flow rate, the instantaneous power of each air compressor, and the status marker data to obtain a power-output neural network model; Train a model according to a preset fully connected neural network algorithm and using the instantaneous power and instantaneous frequency of each air compressor to obtain a power-frequency neural network model; Store the power-output neural network model and the power-frequency neural network model in the cloud.

[0013] Optionally, in the air compressor optimization system based on artificial intelligence and cloud-edge collaboration described in this application, the edge side adjusts the frequency of the air compressor according to the power-output neural network model and the power-frequency neural network model, including: The edge side obtains the current pressure value, current power value, current flow rate, and current status marker data of the air compressor; Perform proportional-integral operation according to the current pressure value and the set pressure value to obtain the system power increase or decrease value, and determine the adjusted power value and the set value of the adjustment times; Input the adjusted power value and the current status marker data into the power-output neural network model for processing to obtain the adjusted flow rate; Calculate the difference between the adjusted flow rate and the current flow rate, and select the target air compressor according to the difference; Input the adjusted power value into the power-frequency neural network model for processing to obtain the adjusted frequency value, set the target air compressor frequency converter to the adjusted frequency value, and synchronously update the current adjustment times; Compare the current adjustment times with the set value of the adjustment times. If the current adjustment times are less than the set value of the adjustment times, continue to repeat the adjustment.

[0014] Optionally, in the air compressor optimization system based on artificial intelligence and cloud-edge collaboration described in this application, use a preset long short-term memory artificial neural network algorithm to construct a demand prediction model and store it in the cloud. The edge side predicts the current demand of the air compressor unit according to the demand prediction model to obtain the expected demand, including: Obtain the average demand of the air compressor, the average flow rate, and the instantaneous pressure within a preset time period; Based on a preset long short-term memory artificial neural network algorithm, and perform model training according to the average flow rate, instantaneous pressure, and average demand within a preset time period to obtain a demand prediction model and store it in the cloud; The edge side obtains the average flow rate and the instantaneous pressure of the air compressor unit within the current preset time period; Input the average flow rate and the instantaneous pressure within the current preset time period into the preset demand prediction model for processing to obtain the expected demand.

[0015] As can be seen from the above, the air compressor regulation optimization method and system based on artificial intelligence and cloud-edge collaboration provided by this application model the air compressor performance model using a fully connected neural network (FCNN) in the cloud and model the required amount of compressed air using a long short-term memory neural network (LSTM). At the edge side, frequency conversion optimization and scheduling optimization are performed according to the performance model and demand model synchronized from the cloud to ensure that the air compressor unit is always in the optimal efficiency state under working conditions, achieving the purpose of energy conservation and cost reduction.

[0016] Other features and advantages of this application will be described in the following specification. Moreover, some of them will become obvious from the specification or can be understood by implementing the embodiments of this application. The objectives and other advantages of this application can be achieved and obtained by the structures specifically pointed out in the written specification and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the following will briefly introduce the drawings required to be used in the embodiments of this application. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can also be obtained based on these drawings without creative efforts.

[0018] Figure 1 It is a flowchart of the air compressor regulation optimization method based on artificial intelligence and cloud-edge collaboration provided by the embodiments of this application; Figure 2 It is a flowchart of modeling the air compressor performance model in the air compressor regulation optimization method based on artificial intelligence and cloud-edge collaboration provided by the embodiments of this application; Figure 3 It is a flowchart of frequency adjustment for the air compressor in the air compressor regulation optimization method based on artificial intelligence and cloud-edge collaboration provided by the embodiments of this application; Figure 4 It is a flowchart of obtaining the expected required amount in the air compressor regulation optimization method based on artificial intelligence and cloud-edge collaboration provided by the embodiments of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application usually described and illustrated in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application claimed, but merely represents selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative efforts belong to the scope of protection of the present application.

[0020] It should be noted that similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of the present application, terms such as "first" and "second" are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0021] Please refer to Figure 1 , Figure 1 which is a flowchart of an air compressor optimization method based on artificial intelligence and cloud-edge collaboration in some embodiments of the present application. The air compressor optimization method based on artificial intelligence and cloud-edge collaboration is used in terminal devices such as computers and mobile phone terminals. The air compressor optimization method based on artificial intelligence and cloud-edge collaboration includes the following steps: S11. Use a preset fully connected neural network algorithm to model the air compressor performance model, obtain a power-output neural network model and a power-frequency neural network model, and store them in the cloud; S12. The edge side adjusts the frequency of the air compressor according to the power-output neural network model and the power-frequency neural network model; S13. Use a preset long short-term memory artificial neural network algorithm to construct a demand prediction model and store it in the cloud. The edge side predicts the current demand of the air compressor unit according to the demand prediction model to obtain the expected demand; S14. The edge side calculates the expected total power of the air compressor unit when the air compressor unit reaches the expected demand; S15. The edge side obtains the full-load flow rate of the air compressor, compares the expected demand with the full-load flow rate, and selects the best scheduling optimization plan for the air compressor unit according to the comparison result.

[0022] It should be noted that in this application, a fully connected neural network (FCNN) is used in the cloud to build an air compressor performance model, and a long short-term memory neural network (LSTM) is used to build a compressed air demand model. At the edge side, frequency conversion optimization and scheduling optimization are performed according to the performance model and demand model synchronized from the cloud to ensure that the air compressor unit is always in the optimal efficiency state under working conditions, achieving the purpose of energy conservation and cost reduction.

[0023] Please refer to Figure 2 , Figure 2 is a flowchart for building an air compressor performance model in the air compressor optimization method based on artificial intelligence and cloud-edge collaboration in some embodiments of this application. According to the embodiments of the present invention, the use of a preset fully connected neural network algorithm to build an air compressor performance model, obtaining a power-output neural network model and a power-frequency neural network model, and storing them in the cloud includes: S21. Real-time collect the operating data of the air compressor, including instantaneous power, instantaneous frequency, status flag data, and average flow rate; S22. Based on the preset fully connected neural network algorithm, use the instantaneous power, status flag data, and average flow rate of a single air compressor to initialize the parameters of the preset power-output model to obtain an initialized power-output neural network model; S23. Calculate the total average flow rate of the air compressor unit based on the average flow rates of all air compressors; S24. Train the initialized power-output neural network model according to the total average flow rate, the instantaneous power, and the status flag data of each air compressor to obtain a power-output neural network model; S25. According to the preset fully connected neural network algorithm and using the instantaneous power and instantaneous frequency of each air compressor for model training to obtain a power-frequency neural network model; S26. Store the power-output neural network model and the power-frequency neural network model in the cloud.

[0024] It should be noted that the status flag data represents the start / stop and loading / unloading states as two binary features, running and loaded as 1, stopped or unloaded as 0. The average flow rate is the average flow rate of the air compressor within 5 seconds, and the instantaneous power is input at intervals of 0.5 kW. First, use the data of a single running unit to initialize the model parameters, and then combine all the air compressors in the same unit and use the data of multiple units running simultaneously for model training to obtain a power-output neural network model. Among them, the total average flow rate of the air compressor unit is the sum of the average flow rates of all air compressors within 5 seconds.

[0025] Please refer to Figure 3 , Figure 3It is a flowchart for frequency adjustment of an air compressor in the air compressor optimization method based on artificial intelligence and cloud-edge collaboration in some embodiments of the present application. According to the embodiments of the present invention, the edge side adjusts the frequency of the air compressor according to the power-output neural network model and the power-frequency neural network model, including: S31. The edge side obtains the current pressure value, current power value, current flow rate, and current status marker data of the air compressor; S32. Perform proportional-integral operation according to the current pressure value and the set pressure value to obtain the system power increase or decrease value, and determine the adjusted power value and the set value of the adjustment times; S33. Input the adjusted power value and the current status marker data into the power-output neural network model for processing to obtain the adjusted flow rate; S34. Calculate the difference between the adjusted flow rate and the current flow rate, and select the target air compressor according to the difference; S35. Input the adjusted power value into the power-frequency neural network model for processing to obtain the adjusted frequency value, set the frequency converter of the target air compressor to the adjusted frequency value, and synchronously update the current adjustment times; S36. Compare the current adjustment times with the set value of the adjustment times. If the current adjustment times are less than the set value of the adjustment times, continue to repeat the adjustment.

[0026] It should be noted that the proportional-integral calculation formula is: ; ; Wherein, is the power increase or decrease value at time t, is the proportionality coefficient, is the integral coefficient, is the set pressure value at time t is the current pressure value at time t, is the error, is the integral of the error; If the calculated value is the increased value of the system power, then the sum of the current power value and 0.5KW is used as the adjusted power value. If the calculated value is the decreased value of the system power, then the difference between the current power value and 0.5KW is used as the adjusted power value, and the ratio of the increased or decreased value of the system power to 0.5KW is used as the adjustment times; Calculate the difference between the adjusted flow rate and the current flow rate. If the calculated value is the increased value of the system power, then the air compressor with the largest absolute value of the difference is used as the target air compressor. If the calculated value is the decreased value of the system power, then the air compressor with the smallest absolute value of the difference is used as the target air compressor. That is, if the proportional-integral operation requires an increase in power, then based on the current power, select the air compressor with the largest increase in output when increasing the power by 0.5kW, and increase the frequency of this air compressor to reach the increased power. If the proportional-integral operation requires a decrease in power, then based on the current power, select the air compressor with the smallest decrease in output when reducing the power by 0.5kW, and reduce the frequency of this air compressor to reach the decreased power. Repeat this process multiple times until the increased or decreased power reaches the given value of the increased or decreased power, and obtain the optimized frequencies of each air compressor. Among them, each time an adjustment is made, the current adjustment times is incremented by 1.

[0027] Please refer to Figure 4 , Figure 4 is a flowchart of obtaining the expected demand in the air compressor optimization method based on artificial intelligence and cloud-edge collaboration in some embodiments of the present application. According to an embodiment of the present invention, the method of using the preset long short-term memory artificial neural network algorithm to construct a demand prediction model and store it in the cloud, and the edge side predicts the current demand of the air compressor unit according to the demand prediction model to obtain the expected demand, includes: S41. Obtain the average demand of the air compressor, the average flow rate and the instantaneous pressure within a preset time period; S42. Based on the preset long short-term memory artificial neural network algorithm, and according to the average flow rate, instantaneous pressure and average demand within the preset time period, perform model training to obtain a demand prediction model and store it in the cloud; S43. The edge side obtains the average flow rate and the instantaneous pressure of the air compressor unit within the current preset time period; S44. Input the average flow rate and the instantaneous pressure within the current preset time period into the preset demand prediction model for processing to obtain the expected demand.

[0028] It should be noted that the average flow rate and the instantaneous pressure of the air compressor within the preset time period refer to the average flow rate of 15 seconds in the first 60 minutes and the instantaneous pressure in the first 60 minutes at the moment of demand prediction, and the average demand refers to the average flow rate demand in the next 5 minutes at the moment of demand prediction.

[0029] According to an embodiment of the present invention, the edge side calculates the expected total power of the air compressor unit when it reaches the expected demand, including: The edge device obtains the efficiency corresponding to each air compressor at different powers, and selects the power value with the highest efficiency and the average flow rate as the optimal power value and the optimal output value; Sum up the optimal power values of all air compressors to obtain the optimal total power of the air compressor unit, and sum up the optimal output values of all air compressors to obtain the optimal total output of the air compressor unit; Determine the adjusted power value according to the comparison result between the expected demand and the optimal total output; Input the adjusted power value and the current status marker data into the power-output neural network model for processing to obtain the adjusted flow rate; Calculate the difference between the adjusted flow rate and the current flow rate, and select the target air compressor according to the difference; Input the adjusted power value into the power-frequency neural network model for processing to obtain the adjusted frequency value, set the frequency converter of the target air compressor to the adjusted frequency value, and calculate the adjusted average flow rate of the air compressor unit according to the adjusted flow rate of the target air compressor; Compare the adjusted average flow rate of the air compressor unit with the expected demand. If they are consistent, use the total power of the air compressor unit at this time as the expected total power of the air compressor unit; If they are inconsistent, continue to repeat the adjustment until the adjusted average flow rate of the air compressor unit reaches the expected demand, and use the total power of the air compressor unit at this time as the expected total power of the air compressor unit.

[0030] It should be noted that the efficiency corresponding to each power is calculated from the performance curves of each air compressor unit, where efficiency = output / energy consumption. Calculate the output and power of the optimal efficiency of each air compressor, and sum up the output and power of all started and loaded air compressors as the output and power of the optimal efficiency of the air compressor unit. If the expected flow rate is larger than the optimal output, select the air compressor with the largest increase in output when increasing the power by 0.5kW for power adjustment. If the expected flow rate is smaller than the optimal output, select the air compressor with the smallest decrease in output when reducing the power by 0.5kW for power adjustment. Repeat multiple times until the output reaches the expected flow rate, and obtain the total power of the unit at this time.

[0031] Among them, compare the expected demand with the optimal total output. If the expected demand is greater than the optimal total output, use the sum of the current power value of each air compressor and 0.5KW as the adjusted power value. If the expected demand is less than the optimal total output, use the difference between the current power value of each air compressor and 0.5KW as the adjusted power value; if the expected demand is greater than the optimal total output, use the air compressor with the largest absolute value of the difference as the target air compressor. If the expected demand is less than the optimal total output, use the air compressor with the smallest absolute value of the difference as the target air compressor.

[0032] According to the embodiments of the present invention, the edge device obtains the full-load flow rate of the air compressor, compares the expected demand with the full-load flow rate, and selects the best scheduling optimization scheme for the air compressor unit according to the comparison result, including: If the full-load flow rate is greater than the expected demand, calculate the expected total power of the air compressor unit with one additional air compressor and two additional air compressors respectively, and select the plan with the lowest expected total power for execution; If the full-load flow rate is less than the expected demand, calculate the expected total power and energy-saving rate of the air compressor unit with one additional air compressor, one less air compressor, and one replaced air compressor respectively, and select the plan with an energy-saving rate greater than 5% and the highest energy-saving rate for execution.

[0033] It should be noted that in this application, it is judged whether the currently running and loaded air compressor satisfies that the full-load output is greater than the expected demand. If it is satisfied, calculate the expected total power of adding one, reducing one, and replacing one respectively, and select the plan with the highest energy-saving rate that satisfies the energy-saving rate greater than 5% and the full-load output greater than the expected demand. Among them, the energy-saving rate can be obtained by calculating the ratio of the difference between the energy consumption before transformation and the energy consumption after transformation to the energy consumption before transformation. If the full-load output is not greater than the expected demand, calculate the expected total power of adding one and adding two respectively, and select the plan with the lowest total power that satisfies the full-load output greater than the expected demand for execution.

[0034] The present invention also discloses an air compressor optimization system based on artificial intelligence and cloud-edge collaboration, including a memory and a processor. The memory stores an air compressor optimization method program based on artificial intelligence and cloud-edge collaboration. When the air compressor optimization method program based on artificial intelligence and cloud-edge collaboration is executed by the processor, the following steps are realized: Use the preset fully connected neural network algorithm to model the air compressor performance model, obtain the power-output neural network model and the power-frequency neural network model, and store them in the cloud; The edge side adjusts the frequency of the air compressor according to the power-output neural network model and the power-frequency neural network model; Use the preset long short-term memory artificial neural network algorithm to construct a demand prediction model and store it in the cloud. The edge side predicts the current demand of the air compressor unit according to the demand prediction model to obtain the expected demand; The edge side calculates the expected total power of the air compressor unit when it reaches the expected demand; The edge side obtains the full-load flow rate of the air compressor, compares the expected demand with the full-load flow rate, and selects the best scheduling optimization plan for the air compressor unit according to the comparison result.

[0035] It should be noted that in this application, the fully connected neural network (FCNN) is used in the cloud to model the air compressor performance model, and the long short-term memory neural network (LSTM) is used to model the compressed air demand. The edge side performs frequency optimization and scheduling optimization according to the performance model and demand model synchronized from the cloud to ensure that the air compressor unit is always in the optimal efficiency state under the working conditions, achieving the purpose of energy saving and cost reduction.

[0036] According to an embodiment of the present invention, the method for modeling the performance model of an air compressor by using a preset fully connected neural network algorithm to obtain a power-output neural network model and a power-frequency neural network model and storing them in the cloud includes: Collect the operating data of the air compressor in real time, including instantaneous power, instantaneous frequency, status marker data, and average flow rate; Based on the preset fully connected neural network algorithm, use the instantaneous power, status marker data, and average flow rate of a single air compressor to initialize the parameters of the preset power-output model to obtain an initialized power-output neural network model; Calculate the total average flow rate of the air compressor unit according to the average flow rate of all air compressors; Train the initialized power-output neural network model according to the total average flow rate, the instantaneous power, and the status marker data of each air compressor to obtain a power-output neural network model; Train a model according to the preset fully connected neural network algorithm and use the instantaneous power and instantaneous frequency of each air compressor to obtain a power-frequency neural network model; Store the power-output neural network model and the power-frequency neural network model in the cloud.

[0037] It should be noted that the status marker data represents the start-stop and loading-unloading states as two binary features, running and loading is 1, stopping or unloading is 0, the average flow rate is the average flow rate of the air compressor within 5 seconds, and the instantaneous power is input at intervals of 0.5 kW. First, use the data of a single running air compressor to initialize the model parameters, and then combine all the air compressors in the same unit and use the data of multiple air compressors running simultaneously to train the model to obtain a power-output neural network model. Among them, the total average flow rate of the air compressor unit is the sum of the average flow rates of all air compressors within 5 seconds.

[0038] According to an embodiment of the present invention, the edge side adjusts the frequency of the air compressor according to the power-output neural network model and the power-frequency neural network model, including: The edge side obtains the current pressure value, current power value, current flow rate, and current status marker data of the air compressor; Perform proportional-integral operation according to the current pressure value and the set pressure value to obtain the system power increase or decrease value, and determine the adjusted power value and the set value of the adjustment times; Input the adjusted power value and the current status marker data into the power-output neural network model for processing to obtain the adjusted flow rate; Calculate the difference between the adjusted flow rate and the current flow rate, and select the target air compressor according to the difference; Input the adjusted power value into the power - frequency neural network model for processing to obtain the adjusted frequency value, set the target air compressor frequency converter to the adjusted frequency value, and synchronously update the current adjustment times. Compare the current adjustment times with the set value of adjustment times. If the current adjustment times are less than the set value of adjustment times, continue to repeat the adjustment.

[0039] It should be noted that the proportional - integral calculation formula is: ; ; Among them, is the power increase - decrease value at time t, is the proportional coefficient, is the integral coefficient, is the set pressure value at time t, is the current pressure value at time t, is the error, is the integral of the error; If the calculated value is the system power increase value, then use the sum of the current power value and 0.5KW as the adjusted power value. If the calculated value is the system power decrease value, then use the difference between the current power value and 0.5KW as the adjusted power value, and use the ratio of the system power increase - decrease value to 0.5KW as the adjustment times; calculate the difference between the adjusted flow rate and the current flow rate. If the calculated value is the system power increase value, then use the air compressor with the largest absolute value of the difference as the target air compressor. If the calculated value is the system power decrease value, then use the air compressor with the smallest absolute value of the difference as the target air compressor. That is, if the proportional - integral operation requires an increase in power, then based on the current power, select the air compressor with the largest increase in output when increasing the power by 0.5kW, and increase the frequency of this air compressor to reach the increased power. If the proportional - integral operation requires a decrease in power, then based on the current power, select the air compressor with the smallest decrease in output when decreasing the power by 0.5kW, and decrease the frequency of this air compressor to reach the decreased power. Repeat multiple times until the increased or decreased power reaches the given power increase - decrease value to obtain the optimized frequencies of each air compressor. Among them, each time an adjustment is made, the current adjustment times are incremented by 1.

[0040] According to the embodiments of the present invention, using the preset long - short - term memory artificial neural network algorithm to construct a demand prediction model and store it in the cloud, the edge side predicts the current demand of the air compressor unit according to the demand prediction model to obtain the expected demand, including: Obtain the average demand of the air compressor, the average flow rate, and the instantaneous pressure within a preset time period; Based on the preset long - short - term memory artificial neural network algorithm, and according to the average flow rate, instantaneous pressure, and average demand within the preset time period, perform model training to obtain the demand prediction model and store it in the cloud; The edge device obtains the average flow rate and instantaneous pressure of the air compressor unit within the current preset time period; Input the average flow rate and instantaneous pressure within the current preset time period into the preset demand prediction model for processing to obtain the expected demand.

[0041] It should be noted that the average flow rate and instantaneous pressure of the air compressor within the preset time period refer to the 15-second average flow rate in the previous 60 minutes and the instantaneous pressure in the previous 60 minutes at the demand prediction moment, and the average demand refers to the average flow rate demand in the next 5 minutes at the demand prediction moment.

[0042] According to an embodiment of the present invention, the edge device calculates the expected total power of the air compressor unit when reaching the expected demand, including: The edge device obtains the efficiency corresponding to each air compressor at different powers, and selects the power value with the highest efficiency and the average flow rate as the optimal power value and the optimal output value; Sum up the optimal power values of all air compressors to obtain the optimal total power of the air compressor unit, and sum up the optimal output values of all air compressors to obtain the optimal total output of the air compressor unit; Determine the adjusted power value according to the comparison result between the expected demand and the optimal total output; Input the adjusted power value and the current status marking data into the power-output neural network model for processing to obtain the adjusted flow rate; Calculate the difference between the adjusted flow rate and the current flow rate, and select the target air compressor according to the difference; Input the adjusted power value into the power-frequency neural network model for processing to obtain the adjusted frequency value, set the frequency converter of the target air compressor to the adjusted frequency value, and calculate the adjusted average flow rate of the air compressor unit according to the adjusted flow rate of the target air compressor; Compare the adjusted average flow rate of the air compressor unit with the expected demand. If they are consistent, use the total power of the air compressor unit at this time as the expected total power of the air compressor unit; If they are inconsistent, continue to repeat the adjustment until the adjusted average flow rate of the air compressor unit reaches the expected demand, and use the total power of the air compressor unit at this time as the expected total power of the air compressor unit.

[0043] It should be noted that calculate the efficiency corresponding to each power from the performance curves of each air compressor unit, where efficiency = output / energy consumption, calculate the output and power of the optimal efficiency of each air compressor, sum up the output and power of all started and loaded air compressors as the output and power of the optimal efficiency of the air compressor unit. If the expected flow rate is larger than the optimal output, select the air compressor with the largest increase in output when increasing the power by 0.5kW for power adjustment. If the expected flow rate is smaller than the optimal output, select the air compressor with the smallest decrease in output when reducing the power by 0.5kW for power adjustment. Repeat multiple times until the output reaches the expected flow rate, and obtain the total power of the unit at this time.

[0044] Among them, the expected demand is compared with the optimal total output. If the expected demand is greater than the optimal total output, the sum of the current power values of each air compressor and 0.5KW is used as the adjusted power value. If the expected demand is less than the optimal total output, the difference between the current power value of each air compressor and 0.5KW is used as the adjusted power value. If the expected demand is greater than the optimal total output, the air compressor with the largest absolute value of the difference is used as the target air compressor. If the expected demand is less than the optimal total output, the air compressor with the smallest absolute value of the difference is used as the target air compressor.

[0045] According to the embodiment of the present invention, the edge end obtains the full-load flow rate of the air compressor, compares the expected demand with the full-load flow rate, and selects the best scheduling optimization scheme for the air compressor unit according to the comparison result, including: If the full-load flow rate is greater than the expected demand, calculate the expected total power of the air compressor unit with one additional air compressor and two additional air compressors respectively, and select the scheme with the lowest expected total power for execution; If the full-load flow rate is less than the expected demand, calculate the expected total power and energy-saving rate of the air compressor unit with one additional air compressor, one less air compressor, and one replaced air compressor respectively, and select the scheme with an energy-saving rate greater than 5% and the highest energy-saving rate for execution.

[0046] It should be noted that in this application, it is judged whether the currently operating and loaded air compressor satisfies that the full-load output is greater than the expected demand. If it is satisfied, calculate the expected total power of adding one, reducing one, and replacing one respectively, and select the scheme with the highest energy-saving rate that satisfies the energy-saving rate greater than 5% and the full-load output greater than the expected demand. Among them, the energy-saving rate can be calculated by the ratio of the difference between the energy consumption before transformation and the energy consumption after transformation to the energy consumption before transformation. If the full-load output is not greater than the expected demand, calculate the expected total power of adding one and adding two respectively, and select the scheme with the lowest total power that satisfies the full-load output greater than the expected demand for execution.

[0047] The air compressor optimization method and system based on artificial intelligence and cloud-edge collaboration disclosed by the present invention model the air compressor performance model by using a fully connected neural network (FCNN) in the cloud, model the compressed air demand by using a long short-term memory neural network (LSTM), and perform frequency conversion optimization and scheduling optimization at the edge end according to the performance model and demand model synchronized from the cloud, ensuring that the air compressor unit is always in the optimal efficiency state under working conditions, achieving the purpose of energy conservation and cost reduction.

[0048] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined, or can be integrated into another system, or some features can be ignored, or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces. The indirect coupling or communication connection of devices or units can be electrical, mechanical, or other forms.

[0049] The units described above as separate components may or may not be physically separated. The components shown as units may or may not be physical units; they can be located in one place or distributed to multiple network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0050] In addition, each functional unit in the embodiments of the present invention can be all integrated in a processing unit, or each unit can be separately used as a unit, or two or more units can be integrated in a unit; the above integrated unit can be implemented in the form of hardware, or in the form of a combination of hardware and software functional units.

Claims

1. An air compressor optimization method based on artificial intelligence and cloud-edge collaboration, characterized in that Including the following steps: Using a preset fully connected neural network algorithm to build an air compressor performance model, obtaining a power-output neural network model and a power-frequency neural network model, and storing them in the cloud; The edge side adjusts the frequency of the air compressor according to the power-output neural network model and the power-frequency neural network model; Using a preset long short-term memory artificial neural network algorithm to build a demand prediction model and storing it in the cloud, and the edge side predicts the current demand of the air compressor unit according to the demand prediction model to obtain the expected demand; The edge side calculates the expected total power of the air compressor unit when it reaches the expected demand; The edge side obtains the full-load flow rate of the air compressor, compares the expected demand with the full-load flow rate, and selects the best scheduling optimization plan for the air compressor unit according to the comparison result.

2. The air compressor optimization method based on artificial intelligence and cloud-edge collaboration according to claim 1, wherein, The step of using a preset fully connected neural network algorithm to build an air compressor performance model, obtaining a power-output neural network model and a power-frequency neural network model, and storing them in the cloud includes: Real-time collecting the operation data of the air compressor, including instantaneous power, instantaneous frequency, status marker data, and average flow rate; Based on the preset fully connected neural network algorithm, using the instantaneous power, status marker data, and average flow rate of a single air compressor to initialize the parameters of the preset power-output model to obtain an initialized power-output neural network model; Calculating the total average flow rate of the air compressor unit according to the average flow rate of all air compressors; Training the initialized power-output neural network model according to the total average flow rate, the instantaneous power, and the status marker data of each air compressor to obtain a power-output neural network model; Training the model according to the preset fully connected neural network algorithm and using the instantaneous power and instantaneous frequency of each air compressor to obtain a power-frequency neural network model; Storing the power-output neural network model and the power-frequency neural network model in the cloud.

3. The air compressor tuning optimization method based on artificial intelligence and cloud-edge collaboration according to claim 2, wherein The step that the edge side adjusts the frequency of the air compressor according to the power-output neural network model and the power-frequency neural network model includes: The edge side obtains the current pressure value, current power value, current flow rate, and current status marker data of the air compressor; Performing proportional-integral operation according to the current pressure value and the set pressure value to obtain the system power increase or decrease value, and determining the adjusted power value and the set value of the adjustment times; Inputting the adjusted power value and the current status marker data into the power-output neural network model for processing to obtain the adjusted flow rate; Calculating the difference between the adjusted flow rate and the current flow rate, and selecting the target air compressor according to the difference; Inputting the adjusted power value into the power-frequency neural network model for processing to obtain the adjusted frequency value, setting the target air compressor frequency converter to the adjusted frequency value, and synchronously updating the current adjustment times; Comparing the current adjustment times with the set value of the adjustment times. If the current adjustment times are less than the set value of the adjustment times, continue to repeat the adjustment.

4. The air compressor optimization method based on artificial intelligence and cloud-edge collaboration according to claim 3, wherein The step of using a preset long short-term memory artificial neural network algorithm to build a demand prediction model and storing it in the cloud, and the edge side predicts the current demand of the air compressor unit according to the demand prediction model to obtain the expected demand includes: Obtain the average demand of the air compressor, as well as the average flow rate and instantaneous pressure within a preset time period; Based on the preset long short-term memory artificial neural network algorithm, and perform model training according to the average flow rate, instantaneous pressure, and average demand within the preset time period, obtain a demand prediction model and store it in the cloud; The edge device obtains the average flow rate and instantaneous pressure of the air compressor unit within the current preset time period; Input the average flow rate and instantaneous pressure within the current preset time period into the preset demand prediction model for processing to obtain the expected demand.

5. The air compressor optimization method based on artificial intelligence and cloud-edge collaboration according to claim 4, characterized in that The edge device calculates the expected total power of the air compressor unit when the air compressor unit reaches the expected demand, including: The edge device obtains the efficiency corresponding to each air compressor at different powers, selects the power value with the highest efficiency and the average flow rate as the optimal power value and the optimal output value; Sum up the optimal power values of all air compressors to obtain the optimal total power of the air compressor unit, and sum up the optimal output values of all air compressors to obtain the optimal total output of the air compressor unit; Determine the adjusted power value according to the comparison result between the expected demand and the optimal total output; Input the adjusted power value and the current status marker data into the power-output neural network model for processing to obtain the adjusted flow rate; Calculate the difference between the adjusted flow rate and the current flow rate, and select the target air compressor according to the difference; Input the adjusted power value into the power-frequency neural network model for processing to obtain the adjusted frequency value, set the frequency converter of the target air compressor to the adjusted frequency value, and calculate the adjusted average flow rate of the air compressor unit according to the adjusted flow rate of the target air compressor; Compare the adjusted average flow rate of the air compressor unit with the expected demand. If they are consistent, use the total power of the air compressor unit at this time as the expected total power of the air compressor unit; If they are inconsistent, continue to repeat the adjustment until the adjusted average flow rate of the air compressor unit reaches the expected demand, and use the total power of the air compressor unit at this time as the expected total power of the air compressor unit.

6. The air compressor optimization method based on artificial intelligence and cloud-edge collaboration according to claim 5, wherein, The edge device obtains the full-load flow rate of the air compressor, compares the expected demand with the full-load flow rate, and selects the best scheduling optimization plan for the air compressor unit according to the comparison result, including: If the full-load flow rate is greater than the expected demand, calculate the expected total power of the air compressor unit with one additional air compressor and two additional air compressors respectively, and select the plan with the lowest expected total power to execute; If the full-load flow rate is less than the expected demand, calculate the expected total power and energy-saving rate of the air compressor unit with one additional air compressor, one less air compressor, and one replaced air compressor respectively, and select the plan with an energy-saving rate greater than 5% and the highest energy-saving rate to execute.

7. An air compressor adjustment and optimization system based on artificial intelligence and cloud-edge collaboration, characterized in that It includes a memory and a processor. The memory stores a program for the air compressor adjustment optimization method based on artificial intelligence and cloud-edge collaboration. When the program for the air compressor adjustment optimization method based on artificial intelligence and cloud-edge collaboration is executed by the processor, the following steps are implemented: Use the preset fully connected neural network algorithm to build an air compressor performance model, obtain a power-output neural network model and a power-frequency neural network model, and store them in the cloud; The edge device adjusts the frequency of the air compressor according to the power-output neural network model and the power-frequency neural network model; Use the preset long short-term memory artificial neural network algorithm to construct a demand prediction model and store it in the cloud. The edge device predicts the current demand of the air compressor unit according to the demand prediction model to obtain the expected demand; The edge device calculates the expected total power of the air compressor unit when it reaches the expected demand; The edge device obtains the full-load flow rate of the air compressor, compares the expected demand with the full-load flow rate, and selects the best scheduling optimization plan for the air compressor unit according to the comparison result.

8. The air compressor optimization system based on artificial intelligence and cloud-edge collaboration according to claim 7, wherein The use of the preset fully connected neural network algorithm to model the performance model of the air compressor, obtaining a power-output neural network model and a power-frequency neural network model, and storing them in the cloud, includes: Real-time collect the operating data of the air compressor, including instantaneous power, instantaneous frequency, status marker data, and average flow rate; Based on the preset fully connected neural network algorithm, use the instantaneous power, status marker data, and average flow rate of a single air compressor to initialize the parameters of the preset power-output model to obtain an initialized power-output neural network model; Calculate the total average flow rate of the air compressor unit according to the average flow rate of all air compressors; Train the initialized power-output neural network model according to the total average flow rate, the instantaneous power, and the status marker data of each air compressor to obtain a power-output neural network model; According to the preset fully connected neural network algorithm and using the instantaneous power and instantaneous frequency of each air compressor for model training to obtain a power-frequency neural network model; Store the power-output neural network model and the power-frequency neural network model in the cloud.

9. The air compressor optimization system based on artificial intelligence and cloud-edge collaboration according to claim 8, wherein The edge device adjusts the frequency of the air compressor according to the power-output neural network model and the power-frequency neural network model, including: The edge device obtains the current pressure value, current power value, current flow rate, and current status marker data of the air compressor; Perform proportional-integral operation according to the current pressure value and the set pressure value to obtain the system power increase or decrease value, and determine the adjusted power value and the set value of the adjustment times; Input the adjusted power value and the current status marker data into the power-output neural network model for processing to obtain the adjusted flow rate; Calculate the difference between the adjusted flow rate and the current flow rate, and select the target air compressor according to the difference; Input the adjusted power value into the power-frequency neural network model for processing to obtain the adjusted frequency value, set the frequency converter of the target air compressor to the adjusted frequency value, and synchronously update the current adjustment times; Compare the current adjustment times with the set value of the adjustment times. If the current adjustment times are less than the set value of the adjustment times, continue to repeat the adjustment.

10. The air compressor optimization system based on artificial intelligence and cloud-edge collaboration according to claim 9, wherein The use of the preset long short-term memory artificial neural network algorithm to construct a demand prediction model and store it in the cloud. The edge device predicts the current demand of the air compressor unit according to the demand prediction model to obtain the expected demand, includes: Obtain the average demand of the air compressor, as well as the average flow rate and instantaneous pressure within a preset time period; Based on the preset long short-term memory artificial neural network algorithm, and perform model training according to the average flow rate, instantaneous pressure, and average demand within the preset time period to obtain a demand prediction model and store it in the cloud; The edge device obtains the average flow rate and instantaneous pressure of the air compressor unit within the current preset time period; Input the average flow velocity and instantaneous pressure within the current preset time period into a preset demand prediction model for processing to obtain the expected demand.

Images