Energy consumption optimization method, system and device for air compressor cluster control system based on genetic algorithm

Abstract

The present invention discloses an energy consumption optimization method, system and device for an air compressor cluster control system based on a genetic algorithm. The method consists of the following steps: a target gas consumption is obtained through a prediction algorithm based on historical data of gas consumption of an air compressor cluster, the target gas consumption is the gas consumption of the air compressor cluster within a predetermined time, a target working state of the air compressor cluster is calculated using a genetic algorithm based on the target gas consumption, when the air compressor cluster executes according to the target working state, the actual gas consumption of the air compressor cluster is measured, and the deviation gas consumption between the actual gas consumption and the target gas consumption is calculated, when the deviation gas consumption is greater than a predetermined value, the remaining gas consumption within the predetermined time is obtained, and the remaining working state of the air compressor cluster is calculated based on the remaining gas consumption, and this is iterated back and forth until the deviation gas consumption is ≤ the predetermined value; the present invention can be combined with the energy efficiency of a single-machine device of the air compressor and the limit on the number of starts and stops to obtain an air compressor cluster control strategy with the lowest energy consumption.

Description

Technical Field

[0001] The present invention belongs to the technical field of air compressor cluster control systems, and in particular relates to an energy consumption optimization method, system and device for an air compressor cluster control system based on a genetic algorithm. Background Art

[0002] Air compressors are basic equipment in industrial development and are responsible for providing energy for factory pneumatic components. However, in the actual application of air compressors, energy consumption is relatively large, accounting for the vast majority of the entire system. Among all large industrial equipment, the power consumption of air compressors can reach about 15%. In order to save energy and improve economic benefits, energy-saving transformation must be carried out on air compressors.

[0003] In large industrial and mining enterprises and manufacturing industries, compressed air is often produced by cluster control of air compressors rather than by a single air compressor. For screw air compressors, the compressors under cluster control can all be non-variable frequency compressors, or a combination of non-variable frequency compressors and variable frequency compressors. The type and exhaust volume of the compressor can be selected according to actual needs. Therefore, under the same air demand, the impact of different air compressor group control methods on energy consumption is crucial.

[0004] The current air compressor group control system does not predict the gas demand, nor does it establish the relationship between the gas demand and the air compressor load. It only calculates the average load of the air compressor that is turned on, and judges whether to increase or decrease the air compressor based on human experience, which is a stress response. In addition, the actual efficiency curve of each air compressor is different. It is difficult to achieve timely response and minimum energy consumption by relying on people to responsively adjust the air compressor switch. The air compressor cluster system is usually a large electricity consumer in the factory, and its control method still faces great challenges in maintaining optimal economy. For example: 1. Although multiple air compressors are connected in parallel to maintain gas production to meet the gas demand, in most cases the control of each air compressor is independent of each other and does not work together. 2. Even if the air compressor cluster with the group control system installed can achieve control synergy, the control method is to respond in a stressful manner according to the load changes of the turned-on air compressor, and predictive optimization cannot be achieved. 3. In the absence of a group control system, when the gas demand changes significantly, it is necessary to manually re-match the relationship between the gas demand and the air compressor load to complete the start-stop combination control of the air compressor. Manual adjustment requires expert experience, and it is difficult to achieve accurate and fast response. It is difficult to maintain the best match between the number of air compressors turned on and the gas demand. When the number of air compressors turned on is insufficient, it will cause air compressor overload and insufficient gas supply. When the number of air compressors turned on is too large, it will cause waste and generate additional costs. Summary of the invention

[0005] The purpose of the present invention is to provide an energy consumption optimization method, system and device for an air compressor cluster control system based on a genetic algorithm, so as to solve the problem that the air compressors in the existing air compressor cluster are independent of each other, cannot be centrally controlled, have poor coordination and require manual adjustment.

[0006] The present invention adopts the following technical solution: a method for optimizing energy consumption of an air compressor cluster control system based on a genetic algorithm, which comprises the following steps:

[0007] The target gas consumption is obtained through the prediction algorithm based on the historical data of the gas consumption of the air compressor cluster. The target gas consumption is the gas consumption of the air compressor cluster within a predetermined time.

[0008] According to the target gas consumption, the target working state of the air compressor cluster is calculated using genetic algorithm based on the mathematical model of non-variable frequency and variable frequency screw air compressors.

[0009] When the air compressor cluster is executed according to the target working state, the actual gas consumption of the air compressor cluster is measured, and the deviation between the actual gas consumption and the target gas consumption is calculated.

[0010] When the deviation gas consumption is greater than the preset value, the prediction algorithm is used to obtain the remaining gas consumption within the preset time.

[0011] According to the remaining gas consumption, based on the mathematical model of non-variable frequency and variable frequency screw air compressors, the genetic algorithm is used to calculate the remaining working state of the air compressor cluster. This iteration is repeated until the deviation gas consumption is ≤ the predetermined value and the iteration stops.

[0012] Furthermore, the prediction algorithm is:

[0013]

[0014] In the formula, y k is the target gas consumption, k is the kth node of the output layer, w jk is the weight between the jth node in the hidden layer and the kth node in the output layer, w ij is the weight between the i-th node in the input layer and the j-th node in the hidden layer, α j is the threshold of the jth node in the hidden layer, β k is the threshold of the kth node in the output layer, m is the number of nodes in the hidden layer, n is the number of nodes in the input layer, is the activation function of the hidden layer; ψ(·) is the activation function of the output layer.

[0015] Furthermore, the genetic algorithm is:

[0016] {E total (type_fsd,t_unload,f_vsd)} min =(E_fsd+E_vsd)min

[0017] In the formula, E total is the total energy consumption of all air compressors in one day, E_fsd is the total energy consumption of all non-variable frequency air compressors in one day; E_vsd is the total energy consumption of all variable frequency air compressors in one day;

[0018] The following two constraints should be met when solving the genetic algorithm:

[0019] ①Satisfy the gas supply volume ≥ gas consumption demand, which can be expressed as:

[0020] g(type_fsd,t_unload,f_vsd)=Q_use-Q_fsd-Q_vsd≤0

[0021] Among them, g is the difference between the gas consumption and the gas output of the air compressor in each stage within the predetermined time; Q_use is the gas consumption; Q_fsd is the gas output of all non-variable frequency air compressors; Q_vsd is the gas output of all variable frequency air compressors;

[0022] ②The frequency of the variable frequency air compressor is within its allowable range.

[0023] Furthermore, the target working state is the start, stop or unloading of each air compressor, and the remaining working state is the start, stop or unloading of each air compressor.

[0024] An energy consumption optimization system for an air compressor cluster control system based on a genetic algorithm, comprising:

[0025] The prediction module is used to obtain the target gas consumption through the prediction algorithm based on the historical data of the gas consumption of the air compressor cluster, and send the target gas consumption. The target gas consumption is the gas consumption of the air compressor cluster within a predetermined time.

[0026] The optimization module is used to receive the target gas consumption, and based on the target gas consumption, calculate the target working state of the air compressor cluster using genetic algorithm based on the mathematical model of non-variable frequency and variable frequency screw air compressors, and send the target working state.

[0027] The execution module is used to receive the target working state and send an execution command to the air compressor cluster according to the received target working state.

[0028] Furthermore, it also includes:

[0029] The measurement module is used to measure the actual gas consumption of the air compressor cluster when the air compressor cluster is executed according to the target working state, and calculate the deviation between the actual gas consumption and the target gas consumption, and send the deviation gas consumption.

[0030] The judgment module is used to receive the deviation gas consumption and judge that when the deviation gas consumption is greater than the preset value, a re-test signal is sent.

[0031] The retest module is used to receive the retest signal and use the prediction algorithm to obtain the remaining gas consumption within the predetermined time.

[0032] The suboptimal module is used to receive the remaining gas consumption and, based on the remaining gas consumption, calculate the remaining working state of the air compressor cluster using a genetic algorithm based on the mathematical model of non-variable frequency and variable frequency screw air compressors, and send the remaining working state to the execution module.

[0033] The execution module is also used to receive the remaining working status and send an execution command to the air compressor cluster according to the received remaining working status.

[0034] Furthermore, the target working state is the start, stop or unloading of each air compressor, and the remaining working state is the start, stop or unloading of each air compressor.

[0035] Furthermore, the historical data of the gas consumption of the air compressor cluster is the daily gas consumption within a year or the average value of the daily gas consumption over multiple years.

[0036] An energy consumption optimization device for air compressor cluster control system based on genetic algorithm, the device is a server, and consists of the following modules:

[0037] The prediction module is used to obtain the target gas consumption through the prediction algorithm based on the historical data of the gas consumption of the air compressor cluster, and send the target gas consumption. The target gas consumption is the gas consumption of the air compressor cluster within a predetermined time.

[0038] The optimization module is used to receive the target gas consumption, and based on the target gas consumption, calculate the target working state of the air compressor cluster using genetic algorithm based on the mathematical model of non-variable frequency and variable frequency screw air compressors, and send the target working state.

[0039] An execution module, used for receiving a target working state and sending an execution command to the air compressor cluster according to the received target working state;

[0040] The measurement module is used to measure the actual gas consumption of the air compressor cluster when the air compressor cluster is executed according to the target working state, and calculate the deviation between the actual gas consumption and the target gas consumption, and send the deviation gas consumption.

[0041] The judgment module is used to receive the deviation gas consumption and judge that when the deviation gas consumption is greater than the preset value, a re-test signal is sent.

[0042] The retest module is used to receive the retest signal and use the prediction algorithm to obtain the remaining gas consumption within the predetermined time.

[0043] The suboptimal module is used to receive the remaining gas consumption and, based on the remaining gas consumption, calculate the remaining working state of the air compressor cluster using a genetic algorithm based on the mathematical model of non-variable frequency and variable frequency screw air compressors, and send the remaining working state to the execution module.

[0044] The execution module is also used to receive the remaining working status and send an execution command to the air compressor cluster according to the received remaining working status.

[0045] A server for an air compressor cluster control system energy consumption optimization method based on a genetic algorithm comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the program, the air compressor cluster control system energy consumption optimization method based on a genetic algorithm is implemented.

[0046] The beneficial effects of the present invention are as follows: the present invention can predict the gas demand in the next day based on the trend characteristics of the time series of gas consumption, the periodic seasonal variation law, the operating status of the air compressor cluster and other information, thereby transforming the traditional stress-type adjustment method into a predictive optimization method; the present invention can establish the relationship between external environmental factors, gas consumption, air compressor load and air compressor energy consumption, calculate the total system energy consumption under different gas consumption demands and air compressor on-off combinations, and obtain the air compressor cluster control strategy with the lowest energy consumption in combination with the single-machine equipment energy efficiency of the air compressor and the limit on the number of starts and stops; compared with the traditional method, this method takes into account more comprehensive influencing factors, takes into account the variability at different times in the future, and saves costs to a greater extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a structural schematic diagram of the present invention;

[0048] Figure 2 This is a flow chart of the energy consumption optimization method of the air compressor cluster control system based on genetic algorithm in the present invention.

[0049] Among them: 1. Prediction module; 2. Optimization module; 3. Execution module; 4. Measurement module; 5. Suboptimal module; 6. Judgment module; 7. Re-test module. DETAILED DESCRIPTION

[0050] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0051] The present invention discloses a method for optimizing energy consumption of an air compressor cluster control system based on a genetic algorithm. Figure 2 As shown, it consists of the following steps:

[0052] Step 1: derive the target gas consumption through a prediction algorithm based on the historical data of the gas consumption of the air compressor cluster. The target gas consumption is the gas consumption of the air compressor cluster within a predetermined time. The historical data of the gas consumption of the air compressor cluster is the daily gas consumption within a year or the average daily gas consumption over multiple years.

[0053] Generally speaking, the external factors that affect the gas consumption of the air compressor cluster are the factory's production plan. However, in different application contexts, there are also other influencing factors, such as weather conditions, temperature, etc.

[0054] The prediction algorithm can be an artificial intelligence algorithm, such as SVM, BP neural network algorithm, LSTM neural network algorithm, fuzzy prediction algorithm, etc. The prediction algorithm used in the present invention is: BP neural network algorithm, the prediction algorithm is:

[0055]

[0056] In the formula, y k is the target gas consumption, k is the kth node of the output layer, w jk is the weight between the jth node in the hidden layer and the kth node in the output layer, w ij is the weight between the i-th node in the input layer and the j-th node in the hidden layer, α j is the threshold of the jth node in the hidden layer, β k is the threshold of the kth node in the output layer, m is the number of nodes in the hidden layer, n is the number of nodes in the input layer, is the activation function of the hidden layer; ψ(·) is the activation function of the output layer.

[0057] The BP neural network consists of an input layer, a hidden layer, and an output layer. It is an artificial neural network algorithm that trains sample data and continuously corrects network weights and thresholds to make the error function decrease along the negative gradient direction and approach the expected output.

[0058] The historical data of gas consumption is used as training samples to train the BP network parameters. The network uses the S-type transfer function By back-propagating the error function Where T i is the expected output, O i For the actual calculation output of the network, the network weights and thresholds are continuously adjusted to make the error function E reach a minimum value, and the network weights and thresholds at this time are saved. The target gas consumption is calculated based on the network weights and thresholds.

[0059] Step 2: According to the target gas consumption, based on the mathematical model of non-variable frequency and variable frequency screw air compressors, the target working state of the air compressor cluster is calculated using a genetic algorithm. The target working state is the start, stop or unloading of each air compressor.

[0060] The genetic algorithm is:

[0061] {E total (type_fsd,t_unload,f_vsd)} min =(E_fsd+E_vsd) min

[0062] In the formula, E total E_fsd is the total energy consumption of all air compressors in one day, E_vsd is the total energy consumption of all variable frequency air compressors in one day.

[0063] The following two constraints should be met when solving the genetic algorithm:

[0064] ① The gas supply is greater than or equal to the gas demand, which can be expressed as:

[0065] g(type_fsd,t_unload,f_vsd)=Q_use-Q_fsd-Q_vsd≤0

[0066] Among them, g is the difference between the gas consumption in each stage within the predetermined time and the gas output of the air compressor, Q_use is the gas consumption, Q_fsd is the gas output of all non-variable frequency air compressors, and Q_vsd is the gas output of all variable frequency air compressors.

[0067] ②The frequency of the variable frequency air compressor is within its allowable range.

[0068] Step 3: When the air compressor cluster is executing according to the target working state, the actual gas consumption of the air compressor cluster is measured, and the deviation between the actual gas consumption and the target gas consumption is calculated. When the deviation gas consumption is greater than the predetermined value, the prediction algorithm is used to obtain the remaining gas consumption within the predetermined time.

[0069] Step 4: According to the remaining gas consumption, based on the mathematical model of non-variable frequency and variable frequency screw air compressors, the genetic algorithm is used to calculate the remaining working state of the air compressor cluster. The remaining working state is the start, stop or unloading of each air compressor. This iteration is repeated until the deviation gas consumption is ≤ the predetermined value and the iteration stops.

[0070] The present invention also discloses an energy consumption optimization system for air compressor cluster control system based on genetic algorithm, such as Figure 1As shown, it includes a prediction module 1, an optimization module 2 and an execution module 3. The prediction module 1 is used to obtain the target gas consumption through a prediction algorithm according to the historical data of the gas consumption of the air compressor cluster, and send the target gas consumption. The target gas consumption is the gas consumption of the air compressor cluster within a predetermined time. The optimization module 2 is used to receive the target gas consumption, and according to the target gas consumption, on the basis of the mathematical model of non-variable frequency and variable frequency screw air compressors, calculate the target working state of the air compressor cluster by using a genetic algorithm, and send the target working state. The execution module 3 is used to receive the target working state, and send an execution command to the air compressor cluster according to the received target working state. The target working state is the start, stop or unload of each air compressor.

[0071] The system of the present invention also includes a measuring module 4, a judging module 6, a re-measurement module 7, and a suboptimal module 5. The measuring module 4 is used to measure the actual gas consumption of the air compressor cluster when the air compressor cluster executes according to the target working state, and calculate the deviation gas consumption between the actual gas consumption and the target gas consumption, and send the deviation gas consumption. The judging module 6 is used to receive the deviation gas consumption, and when the deviation gas consumption is greater than the predetermined value, send a re-measurement signal. The re-measurement module 7 is used to receive the re-measurement signal and use a prediction algorithm to obtain the remaining gas consumption within a predetermined time. The suboptimal module 5 is used to receive the remaining gas consumption, and based on the remaining gas consumption, on the basis of the mathematical model of non-variable frequency and variable frequency screw air compressors, use a genetic algorithm to calculate the remaining working state of the air compressor cluster, and send the remaining working state to the execution module 3. The execution module 3 is also used to receive the remaining working state, and send an execution command to the air compressor cluster according to the received remaining working state. The remaining working state is the start, stop or unloading of each air compressor.

[0072] The present invention also discloses an energy consumption optimization device for an air compressor cluster control system based on a genetic algorithm, the device being a server and consisting of the following modules:

[0073] Prediction module 1 is used to obtain the target gas consumption through the prediction algorithm according to the historical data of the gas consumption of the air compressor cluster, and send the target gas consumption. The target gas consumption is the gas consumption of the air compressor cluster within a predetermined time.

[0074] Optimization module 2 is used to receive the target gas consumption, and based on the target gas consumption, calculate the target working state of the air compressor cluster using genetic algorithm based on the mathematical model of non-variable frequency and variable frequency screw air compressors, and send the target working state.

[0075] Execution module 3, used for receiving the target working state, and sending an execution command to the air compressor cluster according to the received target working state;

[0076] The measuring module 4 is used to measure the actual gas consumption of the air compressor cluster when the air compressor cluster is executed according to the target working state, and calculate the deviation gas consumption between the actual gas consumption and the target gas consumption, and send the deviation gas consumption.

[0077] The judgment module 6 is used to receive the deviation gas consumption and determine that when the deviation gas consumption is greater than a predetermined value, a re-test signal is sent.

[0078] The retest module 7 is used to receive the retest signal and use the prediction algorithm to obtain the remaining gas consumption within a predetermined time.

[0079] The suboptimal module 5 is used to receive the remaining gas consumption, and based on the remaining gas consumption, on the basis of the mathematical models of non-variable frequency and variable frequency screw air compressors, calculates the remaining working state of the air compressor cluster using a genetic algorithm, and sends the remaining working state to the execution module 3, wherein the execution module 3 is also used to receive the remaining working state, and send an execution command to the air compressor cluster according to the received remaining working state.

[0080] The present invention also discloses a server for an air compressor cluster control system energy consumption optimization method based on a genetic algorithm, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the method of steps 1-4 is implemented when the processor executes the program.

[0081] Example 1

[0082] The embodiment includes two non-variable frequency screw air compressors and one variable frequency screw air compressor. The start and stop states type_fsd, unloading time t_unload, and operating frequency f_vsd of the variable frequency air compressor of the two non-variable frequency screw air compressors are selected as optimization parameter variables. The frequency range of the variable frequency screw air compressor is 10HZ~50HZ.

[0083] The optimization goal of air compressor energy-saving operation is to minimize the total energy consumption of the three air compressors under the premise of meeting the gas demand, that is:

[0084] {E total (type_fsd,t_unload,f_vsd)} min =(E_fsd1+E_fsd2+E_vsd) min

[0085] In the formula, E total is the total energy consumption of all air compressors in one day, E_fsd1 is the total energy consumption of No. 1 non-variable air compressor in one day; E_fsd2 is the total energy consumption of No. 2 non-variable air compressor in one day; E_vsd is the total energy consumption of variable frequency air compressor in one day.

[0086] In order to meet the requirement that the gas supply is greater than or equal to the gas demand, it can be expressed as:

[0087] g(type_fsd,t_unload,f_vsd)=Q_use-Q_fsd1-Q_fsd2-Q_vsd≤0

[0088] Among them, g is the difference between the gas consumption in each stage within the scheduled time and the gas production of the air compressor, Q_use is the gas consumption in each stage within the scheduled time, Q_fsd1 is the gas production of No. 1 non-variable air compressor in each stage within the scheduled time, Q_fsd2 is the gas production of No. 2 non-variable air compressor in each stage within the scheduled time, and Q_vsd is the gas production of each stage within the scheduled time of the variable frequency air compressor.

[0089] Calculation of unloading time of variable frequency air compressor:

[0090] For the air storage tank of the air compression system, according to the ideal gas state equation, the pressure inside the air storage tank satisfies the following equation:

[0091]

[0092] In the formula, R is a constant, which is 287, P t is the pressure in the gas tank, V t is the effective volume of the gas tank, T t is the Kelvin temperature of the gas in the gas tank, and m is the mass of the gas in the gas tank.

[0093] In actual use, the gas tank is equivalent to an air intake and exhaust device, and the gas temperature in the gas tank remains basically unchanged. The gas tank undergoes an adiabatic change, and the above formula is transformed and derived to obtain:

[0094]

[0095] In the formula, m i and m0 represent the intake and exhaust mass of the gas storage tank per unit time, kg / min, respectively. Then, the unloading time can be approximated as:

[0096]

[0097] Where Δt is the unloading time, ΔP t It is the difference between the loading pressure and unloading pressure of a non-variable frequency air compressor.

[0098] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for optimizing energy consumption of air compressor cluster control system based on genetic algorithm, characterized in that: It consists of the following steps: The target gas consumption is obtained through a prediction algorithm based on the historical data of the gas consumption of the air compressor cluster. The target gas consumption is the gas consumption of the air compressor cluster within a predetermined time. According to the target gas consumption, the target working state of the air compressor cluster is calculated using genetic algorithm based on the mathematical model of non-variable frequency and variable frequency screw air compressors. When the air compressor cluster is executed according to the target working state, the actual gas consumption of the air compressor cluster is measured, and the deviation between the actual gas consumption and the target gas consumption is calculated. When the deviation gas consumption is greater than the preset value, the prediction algorithm is used to obtain the remaining gas consumption within the preset time. According to the remaining gas consumption, based on the mathematical model of non-variable frequency and variable frequency screw air compressors, the remaining working state of the air compressor cluster is calculated using a genetic algorithm, and the iteration is repeated until the deviation gas consumption is ≤ the predetermined value, and the iteration stops; The prediction algorithm is: In the formula, y k is the target gas consumption, k is the kth node of the output layer, w jk is the weight between the jth node in the hidden layer and the kth node in the output layer, w ij is the weight between the i-th node in the input layer and the j-th node in the hidden layer, α j is the threshold of the jth node in the hidden layer, β k is the threshold of the kth node in the output layer, m is the number of nodes in the hidden layer, n is the number of nodes in the input layer, is the activation function of the hidden layer; ψ(·) is the activation function of the output layer.

2. The method for optimizing energy consumption of an air compressor cluster control system based on a genetic algorithm according to claim 1, characterized in that: The genetic algorithm is: {E total (type_fsd,t_unload,f_vsd)} min =(E_fsd+E_vsd) min In the formula, E total is the total energy consumption of all air compressors in one day, E_fsd is the total energy consumption of all non-variable frequency air compressors in one day; E_vsd is the total energy consumption of all variable frequency air compressors in one day; The following two constraints should be met when solving the genetic algorithm: ①Satisfy the gas supply volume ≥ gas consumption demand, which can be expressed as: g(type_fsd,t_unload,f_vsd)=Q_use-Q_fsd-Q_vsd≤0 Among them, g is the difference between the gas consumption and the gas output of the air compressor in each stage within the predetermined time; Q_use is the gas consumption; Q_fsd is the gas output of all non-variable frequency air compressors; Q_vsd is the gas output of all variable frequency air compressors; ②The frequency of the variable frequency air compressor is within its allowable range; Among them, type_fsd is the start and stop status of the non-variable frequency screw air compressor, t_unload is the unloading time of the non-variable frequency screw air compressor, and f_vsd is the operating frequency of the variable frequency air compressor.

3. The method for optimizing energy consumption of an air compressor cluster control system based on a genetic algorithm according to claim 1 or 2, characterized in that: The target working state is the start, stop or unloading of each air compressor, and the remaining working state is the start, stop or unloading of each air compressor.

4. An energy consumption optimization system for air compressor cluster control system based on genetic algorithm, characterized in that: include: The prediction module (1) is used to obtain a target gas consumption through a prediction algorithm based on the historical data of gas consumption of the air compressor cluster, and send the target gas consumption, wherein the target gas consumption is the gas consumption of the air compressor cluster within a predetermined time. The optimization module (2) is used to receive the target gas consumption, and based on the target gas consumption, calculate the target working state of the air compressor cluster using a genetic algorithm based on the mathematical model of the non-variable frequency and variable frequency screw air compressors, and send the target working state. An execution module (3), used for receiving a target working state and sending an execution command to the air compressor cluster according to the received target working state; The prediction algorithm is: In the formula, y k is the target gas consumption, k is the kth node of the output layer, w jk is the weight between the jth node in the hidden layer and the kth node in the output layer, w ij is the weight between the i-th node in the input layer and the j-th node in the hidden layer, α j is the threshold of the jth node in the hidden layer, β k is the threshold of the kth node in the output layer, m is the number of nodes in the hidden layer, n is the number of nodes in the input layer, is the activation function of the hidden layer; ψ(·) is the activation function of the output layer.

5. The energy consumption optimization system of the air compressor cluster control system based on genetic algorithm according to claim 4 is characterized in that: Also includes: The measuring module (4) is used to measure the actual gas consumption of the air compressor cluster when the air compressor cluster is executed according to the target working state, calculate the deviation gas consumption between the actual gas consumption and the target gas consumption, and send the deviation gas consumption. The judgment module (6) is used to receive the deviation gas consumption and determine that when the deviation gas consumption is greater than a predetermined value, a re-test signal is sent. The retest module (7) is used to receive the retest signal and obtain the remaining gas consumption within a predetermined time using a prediction algorithm. The suboptimal module (5) is used to receive the remaining gas consumption, and based on the remaining gas consumption, calculate the remaining working state of the air compressor cluster using a genetic algorithm based on the mathematical model of the non-variable frequency and variable frequency screw air compressors, and send the remaining working state to the execution module (3). The execution module (3) is further used to receive the remaining working status and send an execution command to the air compressor cluster according to the received remaining working status.

6. The energy consumption optimization system of the air compressor cluster control system based on genetic algorithm according to claim 5 is characterized in that: The target working state is the start, stop or unloading of each air compressor, and the remaining working state is the start, stop or unloading of each air compressor.

7. The energy consumption optimization system of the air compressor cluster control system based on genetic algorithm according to claim 6 is characterized in that: The historical data of the gas consumption of the air compressor cluster is the daily gas consumption within one year or the average value of the daily gas consumption over multiple years.

8. An energy consumption optimization device for air compressor cluster control system based on genetic algorithm, characterized in that: The device is a server, which is composed of the following modules: The prediction module (1) is used to obtain a target gas consumption through a prediction algorithm based on the historical data of gas consumption of the air compressor cluster, and send the target gas consumption, wherein the target gas consumption is the gas consumption of the air compressor cluster within a predetermined time. The optimization module (2) is used to receive the target gas consumption, and based on the target gas consumption, calculate the target working state of the air compressor cluster using a genetic algorithm based on the mathematical model of the non-variable frequency and variable frequency screw air compressors, and send the target working state. An execution module (3), used for receiving a target working state and sending an execution command to the air compressor cluster according to the received target working state; The measuring module (4) is used to measure the actual gas consumption of the air compressor cluster when the air compressor cluster is executed according to the target working state, calculate the deviation gas consumption between the actual gas consumption and the target gas consumption, and send the deviation gas consumption. The judgment module (6) is used to receive the deviation gas consumption and determine that when the deviation gas consumption is greater than a predetermined value, a re-test signal is sent. The retest module (7) is used to receive the retest signal and obtain the remaining gas consumption within a predetermined time using a prediction algorithm. The suboptimal module (5) is used to receive the remaining gas consumption, and based on the remaining gas consumption, calculate the remaining working state of the air compressor cluster using a genetic algorithm based on the mathematical model of the non-variable frequency and variable frequency screw air compressors, and send the remaining working state to the execution module (3). The execution module (3) is further used to receive the remaining working status and send an execution command to the air compressor cluster according to the received remaining working status; The prediction algorithm is: In the formula, y k is the target gas consumption, k is the kth node of the output layer, w jk is the weight between the jth node in the hidden layer and the kth node in the output layer, w ij is the weight between the i-th node in the input layer and the j-th node in the hidden layer, α j is the threshold of the jth node in the hidden layer, β k is the threshold of the kth node in the output layer, m is the number of nodes in the hidden layer, n is the number of nodes in the input layer, is the activation function of the hidden layer; ψ(·) is the activation function of the output layer.

9. A server for an air compressor cluster control system energy consumption optimization method based on a genetic algorithm, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method according to any one of claims 1 to 3 is implemented.

Images