Variable-frequency energy-saving automatic control system of centrifugal machine

Abstract

The invention relates to the technical field of frequency conversion and energy conservation, and particularly discloses a frequency conversion and energy conservation automatic control system of a centrifugal machine, which comprises an energy consumption acquisition module, an energy consumption analysis module, an energy conservation strategy determination module and an energy conservation regulation and control module. The energy consumption acquisition module acquires energy consumption values of different operation stages of the centrifugal machine through a rotating speed sensor and an electric energy meter; the energy consumption analysis module calculates an unstable value and an energy consumption parameter matrix to obtain an operation state matrix, and inputs the operation state matrix into a regulation and control judgment model to judge whether energy consumption is regulated and controlled; the energy-saving strategy determination module obtains the advanced completion time of the working stage of the centrifugal machine by using RBF nerves; the energy-saving regulation and control module determines an average working time parameter and an energy consumption adjustment parameter to control energy consumption output and time. The energy consumption of the centrifugal machine can be accurately analyzed, regulated and controlled, the working efficiency is improved, and the energy consumption is reduced.

Description

Technical Field

[0001] The invention relates to the technical field of frequency conversion and energy saving, and in particular to a frequency conversion and energy saving automatic control system of a centrifuge. Background Art

[0002] In many fields such as industrial production, centrifuge is a commonly used equipment. Its working principle is to use the centrifugal force generated by high-speed rotation to separate different components in the material according to density differences.

[0003] In the existing technology, it is difficult to measure the energy consumption value of each operation stage, and it is impossible to accurately identify the high energy consumption stage, resulting in a lack of targeted energy-saving measures. For example, after the material reaches the target state in the steady speed stage, the centrifuge still runs for a preset time, resulting in energy waste. At the same time, traditional centrifuges cannot effectively evaluate the operating status, it is difficult to judge the invalid working situation, and it is impossible to adjust the working parameters according to the actual situation to optimize the energy consumption output.

[0004] In the prior art, it is not possible to extract characteristic parameters of the centrifuge operation stage, nor is it possible to obtain the time to complete all processes of the centrifuge working stage, thereby obtaining the advanced completion time of the centrifuge working stage. If the average working time parameter and energy consumption adjustment parameter of the centrifuge are determined based on the time to complete all processes of the centrifuge working stage, the energy consumption of the centrifuge is optimized, and energy-saving regulation of the centrifuge operation can be achieved, which is beneficial to reducing the energy consumption of the centrifuge.

[0005] To this end, the present invention provides a frequency conversion energy-saving automatic control system for a centrifuge. Summary of the invention

[0006] The object of the present invention is to provide a frequency conversion energy-saving automatic control system for a centrifuge to solve the above-mentioned background problems.

[0007] The purpose of the present invention can be achieved by the following technical solutions: A first aspect: The present invention provides a frequency conversion energy-saving automatic control system for a centrifuge, comprising: Energy consumption acquisition module: obtains the energy consumption value of the centrifuge at different operating stages; Energy consumption analysis module: According to the energy consumption value of the centrifuge at different operating stages, the non-steady value and energy consumption parameter matrix N of the centrifuge at all operating stages are obtained. ij , the unsteady value of the centrifuge in all operating stages and the energy consumption parameter matrix N ij Conduct numerical analysis to construct the centrifuge operation state matrix Y, input the centrifuge operation state matrix into the control judgment model to determine whether the energy consumption of the centrifuge needs to be regulated; Energy-saving strategy determination module: If the energy consumption of the centrifuge needs to be regulated, obtain the characteristic parameters of the centrifuge operation stage, obtain the time to complete all processes of the centrifuge working stage through the RBF neural network, calculate the advance completion time of the centrifuge working stage, and generate an energy-saving control signal if the advance completion time is greater than or equal to the advance completion time threshold; Energy-saving control module: Based on the energy-saving control signal, the time to complete all processes of the centrifuge working stage is analyzed to determine the average working time parameters and energy consumption adjustment parameters of the centrifuge.

[0008] A second aspect: The present invention provides a frequency conversion energy-saving automatic control method for a centrifuge, comprising: Step 1: Obtain the energy consumption value of the centrifuge at different operating stages; Step 2: According to the energy consumption value of the centrifuge at different operating stages, obtain the non-steady value and energy consumption parameter matrix N of the centrifuge at all operating stages ij , the unsteady value of the centrifuge in all operating stages and the energy consumption parameter matrix N ij Conduct numerical analysis to construct the centrifuge operation state matrix Y, input the centrifuge operation state matrix into the control judgment model to determine whether the energy consumption of the centrifuge needs to be regulated; Step 3: If the energy consumption of the centrifuge needs to be regulated, obtain the characteristic parameters of the centrifuge operation stage, obtain the time to complete all processes of the centrifuge working stage through the RBF neural network, calculate the advanced completion time of the centrifuge working stage, and generate an energy-saving control signal if the advanced completion time is greater than or equal to the advanced completion time threshold; Step 4: Based on the energy-saving control signal, the time to complete all processes of the centrifuge working stage is analyzed to determine the average working time parameters and energy consumption adjustment parameters of the centrifuge.

[0009] Beneficial effects of the present invention: The centrifuge speed is collected in real time, and each operation stage such as acceleration, steady speed, and deceleration is accurately identified according to the speed range of different intervals. At the same time, the energy consumption value of each stage is accurately measured, which provides a detailed and accurate data basis for subsequent analysis. The energy consumption analysis module constructs an energy consumption parameter matrix based on the acquired energy consumption value, calculates the non-steady value of the centrifuge in all operation stages, and then obtains the operation state matrix. It is input into the control judgment model trained by the neural network model. The model can judge whether energy consumption needs to be regulated by learning the training samples containing the operation state matrix and whether to regulate the mark.

[0010] The energy-saving strategy determination module uses RBF neural network to collect characteristic parameters such as centrifuge speed, energy consumption and centrifugal force in real time, builds a classifier and uses genetic algorithm training to calculate the advance completion time of the centrifuge working stage. By comparing it with the preset advance completion time threshold, it can be determined whether the centrifuge completes the centrifugal work before the process standard, thereby identifying whether the centrifuge has invalid work. When it is determined that the centrifuge has an advance completion situation, the average working time parameter and energy consumption adjustment parameter are calculated, and the centrifugal extraction completion time is used as the time for the subsequent preset centrifuge working stage, which effectively reduces energy consumption and is conducive to optimizing the energy consumption of the centrifuge. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The present invention will be further described below in conjunction with the accompanying drawings.

[0012] Figure 1 It is a module diagram of a frequency conversion energy-saving automatic control system of a centrifuge of the present invention; Figure 2 It is a flow chart of a frequency conversion energy-saving automatic control system method of a centrifuge in the present invention; DETAILED DESCRIPTION

[0013] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Embodiment 1

[0014] See also Figure 1 As shown, a frequency conversion energy-saving automatic control system for a centrifuge includes the following modules: Energy consumption acquisition module: obtains the energy consumption value of the centrifuge at different operating stages; In some embodiments, a speed sensor is provided on the centrifuge, and the speed sensor is used to collect the speed of the centrifuge in real time. Different speed ranges are set corresponding to different operation stages of the centrifuge. The collected speed of the centrifuge is compared with the speed ranges of different intervals to identify the current working stage of the centrifuge according to the speed of the centrifuge. At the same time, an electric energy meter is also provided on the centrifuge to measure the energy consumption value of the centrifuge in different operation stages. It needs to be explained that the different operating stages of the centrifuge mainly include: Acceleration stage: When starting to work, the centrifuge drives the rotor to rotate at high speed through electromechanical drive, gradually increasing the angular velocity, so that the sample generates centrifugal force in the container; Steady speed stage: When the centrifuge reaches the set speed, it enters the steady speed stage. In this stage, the liquid in the sample is affected by the centrifugal force and moves to the outside of the centrifuge; Deceleration stage: When the separation process is completed, the centrifuge begins to decelerate; during deceleration, the centrifugal force gradually decreases; The acceleration stage, steady speed stage, and deceleration stage correspond to different speed ranges [0, A], [Aa, A+a], [A, 0], respectively, where A is the preset speed value, a is the speed error, and the value of A is determined by different models of centrifuges and can be 3000 rpm; Energy consumption analysis module: According to the energy consumption value of the centrifuge at different operating stages, the non-steady value and energy consumption parameter matrix N of the centrifuge at all operating stages are obtained. ij , the unsteady value of the centrifuge in all operating stages and the energy consumption parameter matrix N ij Conduct numerical analysis to construct the centrifuge operation state matrix Y, input the centrifuge operation state matrix into the control judgment model to determine whether the energy consumption of the centrifuge needs to be regulated; In some embodiments, the energy consumption values ​​of the centrifuge in different operation stages are obtained, wherein each operation stage is provided with the same number of collection nodes; the energy consumption values ​​collected by the centrifuge in the acceleration stage, the steady speed stage, and the deceleration stage are used to construct an energy consumption parameter matrix N ij , marked in the following ways: ; Among them, N ij Where i represents the different operation stages of the centrifuge, i≤3, 1 means the centrifuge is in the acceleration stage, 2 means the centrifuge is in the steady speed stage, 3 means the centrifuge is in the deceleration stage, N ij Where j represents the number of acquisition nodes at different operation stages; Obtain the unsteady value of the centrifuge at all operating stages, and compare the unsteady value of the centrifuge at all operating stages with the energy consumption parameter matrix N ij Numerical analysis is performed to construct the centrifuge operating state matrix Y. The calculation process is as follows: ; Among them, Y represents the operating state matrix of the centrifuge, and XF represents the unsteady value of the centrifuge in all operating stages; Inputting the operation state matrix of the centrifuge into the control judgment model to judge whether the energy consumption of the centrifuge needs to be regulated. If it exceeds a preset threshold, an energy consumption regulation signal of the centrifuge is generated; otherwise, a normal energy consumption signal of the centrifuge is generated; It should be noted that when the operation state matrix of the centrifuge is input into the control judgment model, in the control judgment model constructed by the input layer, hidden layer, and output layer, the Sigmod activation function of the output layer linearly combines the calculation results of the previous layers and outputs the control probability value. If the control probability value is higher than the preset threshold, a centrifuge energy consumption control signal is generated, otherwise a normal energy consumption signal of the centrifuge is generated; The control judgment model is constructed in the following way: Constructing a neural network model; inputting the regulated training sample data into the neural network model for training; each training sample data includes: an operation state matrix of the centrifuge and a mark indicating whether it is regulated; judging whether the trained model meets the preset conditions; When the trained model meets the preset conditions, the trained model is determined to be a control judgment model.

[0015] The method for determining whether the centrifuge's operating status is regulated is as follows: The non-steady value XF and energy consumption parameter matrix N of the centrifuge in all operating stages ij Perform product processing on each element in to obtain the energy consumption value of the element; The energy consumption value of each element of the energy consumption parameter matrix is ​​summed up to obtain the energy consumption identification value; Compare the energy consumption identification value with the energy consumption identification threshold; If the energy consumption identification value is higher than the energy consumption identification threshold, the operation state matrix is ​​marked as having a regulation identification; If the energy consumption mark value is lower than the energy consumption mark threshold, no processing will be performed; More specifically, the process of obtaining the non-steady state of the centrifuge at all operating stages is as follows: Obtain the energy consumption value of each collection node in all operation stages of the centrifuge. First, compare the energy consumption value of each collection node with the process standard parameter, extract the energy consumption value of all collection nodes that is greater than the process standard parameter, and record it as the excess energy consumption data sample; Divide the excessive energy consumption data samples into m*n level intervals according to the energy consumption values; It should be explained that: m represents the number of grade areas, n represents the size of the grade area; obtain the sample data in each grade interval, and calculate the unstable value of the centrifuge in all operating stages through the formula; the formula is as follows: ; Where XF represents the unstable value of the centrifuge in all operating stages, a=1, 2, ...m, Sm represents the number of excessive energy consumption data samples in the mth level area, and Z m Represents the total number of excessive energy consumption data samples, It represents the proportion of the number of excessive energy consumption data samples in the m-th level area to the total number of samples, that is, the weight coefficient of the excessive energy consumption data samples in the m-th level area, D m represents the energy consumption variance value of the excessive energy consumption data sample in the m-th level area; For example: if the excessive energy consumption data samples are divided into two level intervals according to the energy consumption value, recorded as high volatility samples and low volatility samples, the number of high volatility samples and low volatility samples is obtained: Analysis of high-volatility samples: Calculate the proportion of high-volatility samples in the total number of samples to obtain the high-volatility sample weight coefficient, then calculate the energy consumption square value of all high-volatility samples, and multiply the high-volatility sample weight coefficient with the energy consumption square value of all high-volatility samples to obtain the high-volatility sample coefficient; Analysis of low-volatility samples: Calculate the proportion of low-volatility samples in the total number of samples to obtain the low-volatility sample weight coefficient, then calculate the energy consumption square value of all low-volatility samples, and multiply the low-volatility sample weight coefficient with the energy consumption square value of all low-volatility samples to obtain the low-volatility sample coefficient; The coefficients of high-fluctuation samples and low-fluctuation samples are averaged to obtain the unstable values ​​of the centrifuge in all operating stages; The technical solution of this embodiment is: obtaining the energy consumption value of the centrifuge at different operating stages; according to the energy consumption value of the centrifuge at different operating stages, obtaining the non-stable value and energy consumption parameter matrix N of the centrifuge at all operating stages ij , the unsteady value of the centrifuge in all operating stages and the energy consumption parameter matrix N ij Numerical analysis is performed to construct an operating state matrix Y of the centrifuge, and the operating state matrix of the centrifuge is input into a control judgment model to determine whether the energy consumption of the centrifuge needs to be regulated. The present invention analyzes and evaluates the energy consumption of the centrifuge at each operating stage to achieve an evaluation based on the comprehensive degree of energy consumption data of the centrifuge and abnormal energy consumption fluctuations. Embodiment 2

[0016] like Figure 1 As shown, a frequency conversion energy-saving automatic control system for a centrifuge also includes the following modules: Energy-saving strategy determination module: If the energy consumption of the centrifuge needs to be regulated, obtain the characteristic parameters of the centrifuge operation stage, obtain the time to complete all processes of the centrifuge working stage through the RBF neural network, calculate the advance completion time of the centrifuge working stage, and generate an energy-saving control signal if the advance completion time is greater than or equal to the advance completion time threshold; Among them, the working stage of the centrifuge is mainly the steady speed stage; In some embodiments, the speed sensor is used to collect the speed of the centrifuge in the steady speed stage in real time, the energy consumption of the centrifuge is measured in real time by the electric energy meter, and the centrifugal force of the centrifuge is measured in real time by the centrifugal force meter; the characteristic parameters of the data values ​​in the operation stage of the centrifuge are extracted, specifically: , where n=1,2,3,t b1 Expressed as centrifuge speed, t b2 Expressed as the total energy consumption of the centrifuge, t b3 Expressed as the centrifugal force of the centrifuge; The classifier is constructed by using RBF neural network to identify the centrifuge state according to the characteristic parameters. The RBF neural network is trained by genetic algorithm and the characteristic parameter set at the bth moment is defined as As the input variable of the RBF neural network, the actual output of the RBF neural network is , its fitness function H is: ; Among them, a1 is 0.54, a2 is 0.36, g is the number of hidden neurons, f is the number of neurons in the input layer, e is the number of input samples, S bn is the actual output of the RBF neural network, y bk is the parameter value of the operating state of the centrifuge working stage, k is the operating state value of the centrifuge working stage, and the value of k is 1, 2, 3, 4; y bk The specific parameter values ​​of the operating status of the centrifuge working stage are: ; Among them, y b1 For the beginning of the centrifuge working phase, y b2 To complete a quarter of the centrifuge working phase, y b3 To complete half of the centrifuge working phase, y b4 To complete the entire process of the centrifuge working stage; Calculate the minimum root mean square error between the actual output of the RBF neural network and the operating state of the centrifuge working stage to determine the entire process of the centrifuge completing the centrifuge working stage; The time for completing all processes of the centrifuge working stage is obtained by a timer, and the difference between the preset centrifuge working stage time and the time for completing all processes of the centrifuge working stage is calculated to obtain the advanced completion time; If the ahead finish time is compared with the ahead finish time threshold: If the advanced completion time is greater than or equal to the advanced completion time threshold, it means that the centrifuge completes the centrifugal work before the process standard, but the centrifuge continues to work according to the preset process time, and an energy-saving control signal is generated; If the advanced completion time is less than the advanced completion time threshold, it means that the centrifuge has not completed the centrifugation work before the process standard, and a centrifuge inspection signal is generated; It should be noted that the advanced completion time threshold is preset by those skilled in the art and its value is , where TB is the time of the preset centrifuge working stage, and X can be greater than or equal to 8; When the energy-saving control signal is obtained, it can be determined that the current operation for energy saving of the centrifuge is: adjusting the running time of the centrifuge; When the centrifuge inspection signal is received, arrange for technicians to check the centrifuge; Energy-saving control module: Based on the energy-saving control signal, the time to complete all processes of the centrifuge working stage is analyzed to determine the average working time parameters and energy consumption adjustment parameters of the centrifuge; In some embodiments, the time of all processes of the centrifuge working stage is obtained, marked as the centrifugal extraction completion time, the over-energy consumption data samples of the centrifugal extraction completion time are extracted, and the over-energy consumption data samples of the centrifugal extraction completion time are summed to obtain a first energy consumption output value, and then the non-over-energy consumption data samples of the centrifugal extraction completion time are extracted, and the non-over-energy consumption data samples of the centrifugal extraction completion time are summed to obtain a second energy consumption output value; By formula , calculate and obtain the energy consumption adjustment parameter PT; wherein P1 is the first energy consumption output value, P2 is the second energy consumption output value, and TQ is the centrifugal extraction completion time; The centrifugal extraction completion time is then used as the time of the subsequent preset centrifuge working stage of the centrifuge; The average working time parameters and energy consumption adjustment parameters of the centrifuge are obtained and sent to the energy supply end of the centrifuge to control the energy consumption output and output time of the centrifuge; The technical solution of the embodiment of the present invention is as follows: characteristic parameters of the centrifuge operation stage are obtained, the time for completing all processes of the centrifuge working stage is obtained through an RBF neural network, and the advanced completion time of the centrifuge working stage is calculated; based on the time for completing all processes of the centrifuge working stage, the average working time parameter and energy consumption adjustment parameter of the centrifuge are determined; the present invention determines the factors of abnormal energy consumption, judges whether the centrifuge has invalid work, and determines the average working time parameter and energy consumption adjustment parameter of the centrifuge based on the invalid work of the centrifuge, so as to realize energy-saving regulation of the work of the centrifuge, which can not only improve the working time efficiency of the centrifuge, but also effectively reduce the energy consumption of the centrifuge. Embodiment 3

[0017] like Figure 2As shown, the present embodiment provides a frequency conversion energy-saving automatic control method for a centrifuge, comprising the following steps: Step 1, obtaining the energy consumption value of the centrifuge at different operating stages; Step 2: According to the energy consumption value of the centrifuge at different operating stages, obtain the non-steady value and energy consumption parameter matrix N of the centrifuge at all operating stages ij , the unsteady value of the centrifuge in all operating stages and the energy consumption parameter matrix N ij Conduct numerical analysis to construct the centrifuge operation state matrix Y, input the centrifuge operation state matrix into the control judgment model to determine whether the energy consumption of the centrifuge needs to be regulated; Step 3: If the energy consumption of the centrifuge needs to be regulated, obtain the characteristic parameters of the centrifuge operation stage, obtain the time to complete all processes of the centrifuge working stage through the RBF neural network, calculate the advanced completion time of the centrifuge working stage, and generate an energy-saving control signal if the advanced completion time is greater than or equal to the advanced completion time threshold; Step 4: Based on the energy-saving control signal, the time to complete all processes of the centrifuge working stage is analyzed to determine the average working time parameter and energy consumption adjustment parameter of the centrifuge; The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A frequency conversion energy-saving automatic control system for a centrifuge, characterized in that: include: Energy consumption acquisition module: obtains the energy consumption value of the centrifuge at different operating stages; Energy consumption analysis module: According to the energy consumption value of the centrifuge at different operating stages, the non-steady value and energy consumption parameter matrix N of the centrifuge at all operating stages are obtained. ij , the unsteady value of the centrifuge in all operating stages and the energy consumption parameter matrix N ij Conduct numerical analysis to construct the centrifuge operation state matrix Y, input the centrifuge operation state matrix into the control judgment model to determine whether the energy consumption of the centrifuge needs to be regulated; Energy-saving strategy determination module: If the energy consumption of the centrifuge needs to be regulated, obtain the characteristic parameters of the centrifuge operation stage, obtain the time to complete all processes of the centrifuge working stage through the RBF neural network, calculate the advance completion time of the centrifuge working stage, and generate an energy-saving control signal if the advance completion time is greater than or equal to the advance completion time threshold; Energy-saving control module: Based on the energy-saving control signal, the time to complete all processes of the centrifuge working stage is analyzed to determine the average working time parameters and energy consumption adjustment parameters of the centrifuge.

2. The frequency conversion energy-saving automatic control system of a centrifuge according to claim 1, characterized in that: The method for determining whether the energy consumption of the centrifuge needs to be regulated is: The operation status matrix of the centrifuge is input into the control judgment model to determine whether the energy consumption of the centrifuge needs to be regulated. If it exceeds a preset threshold, the energy consumption of the centrifuge needs to be regulated.

3. The frequency conversion energy-saving automatic control system of a centrifuge according to claim 2, characterized in that: The construction method of the regulation and control judgment model is as follows: Constructing a neural network model; inputting the regulated training sample data into the neural network model for training; Each training sample data includes: the centrifuge operation state matrix Y and a mark indicating whether it is regulated; judging whether the trained model meets the preset conditions; When the trained model meets the preset conditions, the trained model is determined to be a control judgment model.

4. A frequency conversion energy-saving automatic control system for a centrifuge according to claim 3, characterized in that: The method for obtaining the mark indicating whether to regulate is as follows: The non-steady value XF and energy consumption parameter matrix N of the centrifuge in all operating stages ij Perform product processing on each element in to obtain the energy consumption value of the element; The energy consumption value of each element of the energy consumption parameter matrix is ​​summed up to obtain the energy consumption identification value; Compare the energy consumption identification value with the energy consumption identification threshold; If the energy consumption identification value is higher than the energy consumption identification threshold, the operation state matrix is ​​marked as having a regulation identification.

5. The frequency conversion energy-saving automatic control system of a centrifuge according to claim 4, characterized in that: The non-stable value XF is obtained as follows: Obtain the energy consumption value of each collection node in all operation stages of the centrifuge. First, compare the energy consumption value of each collection node with the process standard parameter, extract the energy consumption value of all collection nodes that is greater than the process standard parameter, and record it as the excess energy consumption data sample; Divide the excessive energy consumption data samples into m*n level intervals according to the energy consumption values; Among them, m represents the number of level regions, and n represents the size of the level region; Obtain sample data within each level interval and calculate the unstable value of the centrifuge in all operating stages through the formula; By formula: Obtain the non-steady-state value XF of the centrifuge at all operating stages; Where a=1, 2, ..., m, Sm represents the number of excessive energy consumption data samples in the mth level area, Zm represents the total number of excessive energy consumption data samples, It represents the proportion of the number of excessive energy consumption data samples in the m-th level area to the total number of samples, that is, the weight coefficient of the excessive energy consumption data samples in the m-th level area. Dm represents the energy consumption variance value of the excessive energy consumption data samples in the m-th level area.

6. The frequency conversion energy-saving automatic control system of a centrifuge according to claim 1, characterized in that: The operation state matrix Y is constructed as follows: The energy consumption values ​​of the centrifuge in different operation stages are obtained, where each operation stage is set with the same number of collection nodes; the energy consumption values ​​collected by the centrifuge in the acceleration stage, steady speed stage, and deceleration stage are used to construct the energy consumption parameter matrix N ij , marked in the following ways: ； Among them, N ij Where i represents the different operation stages of the centrifuge, i≤3, 1 means the centrifuge is in the acceleration stage, 2 means the centrifuge is in the steady speed stage, 3 means the centrifuge is in the deceleration stage, N ij Where j represents the number of acquisition nodes at different operation stages; Obtain the unsteady value of the centrifuge at all operating stages, and compare the unsteady value of the centrifuge at all operating stages with the energy consumption parameter matrix N ij Numerical analysis is performed to construct the centrifuge operating state matrix Y. The calculation process is as follows: ； Among them, Y represents the operating state matrix of the centrifuge, and XF represents the unsteady value of the centrifuge in all operating stages.

7. The frequency conversion energy-saving automatic control system of a centrifuge according to claim 1, characterized in that: The method for obtaining the advanced completion time is as follows: Calculate the minimum root mean square error between the actual output of the RBF neural network and the operating state of the centrifuge working stage to determine the entire process of the centrifuge completing the centrifuge working stage; The time to complete all processes of the centrifuge working stage is obtained through the timer, and the difference between the preset centrifuge working stage time and the time to complete all processes of the centrifuge working stage is calculated to obtain the advanced completion time.

8. The frequency conversion energy-saving automatic control system of a centrifuge according to claim 1, characterized in that: The RBF neural network is constructed as follows: The RBF neural network is used to construct a classifier for identifying the centrifuge state according to the characteristic parameters. The RBF neural network is trained using a genetic algorithm and the characteristic parameter set at the bth moment is defined as ; Among them, the characteristic parameter set at time b is used as the input variable of the RBF neural network, and the actual output of the RBF neural network is , its fitness function H is: ； Among them, a1 is 0.54, a2 is 0.36, g is the number of hidden neurons, f is the number of neurons in the input layer, e is the number of input samples, S bn is the actual output of the RBF neural network, y bk is the parameter value of the operating state of the centrifuge working stage, k is the operating state value of the centrifuge working stage, and the value of k is 1, 2, 3, 4; y bk The specific parameter values ​​of the operating status of the centrifuge working stage are: ； Among them, y b1 For the beginning of the centrifuge working phase, y b2 To complete a quarter of the centrifuge working phase, y b3 To complete half of the centrifuge working phase, y b4 To complete the entire process of the centrifuge working stage.

9. A frequency conversion energy-saving automatic control system for a centrifuge according to claim 8, characterized in that: The characteristic parameters are obtained as follows: The speed sensor is used to collect the centrifuge speed in the steady speed stage in real time, the energy consumption of the centrifuge is measured in real time by the electric energy meter, and the centrifugal force of the centrifuge is measured in real time by the centrifugal force meter; the characteristic parameters of the data values ​​in the centrifuge operation stage are extracted, specifically: ； Where n=1,2,3,t b1 Expressed as centrifuge speed, t b2 Expressed as the total energy consumption of the centrifuge, t b3 Expressed as the centrifugal force of the centrifuge.

10. The frequency conversion energy-saving automatic control system of a centrifuge according to claim 1, characterized in that: The energy consumption adjustment parameter is obtained in the following manner: By formula , calculate and obtain the energy consumption adjustment parameter PT; wherein P1 is the first energy consumption output value, P2 is the second energy consumption output value, and TQ is the centrifugal extraction completion time; The centrifugal extraction completion time is then used as the time of the subsequent preset centrifuge working stage of the centrifuge; The average working time parameters and energy consumption adjustment parameters of the centrifuge are obtained and sent to the energy supply end of the centrifuge to control the energy consumption output and output time of the centrifuge.

Images