A Model Predictive Control System and Method for a Medicine Calcining Machine Based on the Entropy Weight Method

Abstract

The present invention discloses a model predictive control system and method for a medicine calcining machine based on the entropy weight method. The method includes constructing a nonlinear model based on the quality data, temperature data, and real-time power of the electrothermal machine during the medicine processing process, constructing a system state space equation, linearizing it, and designing a state observer to observe the value of the first derivative of the real-time temperature of the medicine; discretizing the system state space equation to construct a discretized system; for the discretized system, using the rolling optimization method to predict the system state for a future period of time; designing a quadratic control law, and using the entropy weight method to obtain the weight coefficients of the system state and the weight coefficients of the input variables in the quadratic control law; substituting the predicted system state for a future period of time, setting constraint conditions, solving the system input variable sequence, and taking the first input variable in the solved sequence as the power of the electrothermal machine at the next moment and applying it to the medicine calcining machine.

Description

Technical Field

[0001] The present invention relates to the fields of system control and pharmaceuticals, and particularly to a model predictive control system and method for a medicine calcining machine based on the entropy weight method. Background Art

[0002] In the pharmaceutical industry, especially in traditional Chinese medicine production, temperature is an important environmental factor affecting the efficacy of medicines and the quality of medicinal materials. Too high or too low temperature will have an adverse impact on the quality of Chinese medicinal materials. However, there are a large number of equipment and technological processes that need to control temperature in the pharmaceutical industry, such as medicine calcining machines, dryers, reaction kettles, etc.

[0003] Traditional methods for controlling the temperature of medicinal materials are to use constant temperature equipment or manual adjustment. These methods have problems such as inaccurate temperature control, slow response, and high energy consumption. A more advanced control method is PID control. However, it lacks the ability to monitor the temperature distribution in real time and make dynamic adjustments, and these methods are difficult to adjust parameters, sensitive to interference, and have performance limitations. In addition, the PID controller may be sensitive to noise and external disturbances, resulting in the instability of the system output when facing high-frequency noise. Summary of the Invention

[0004] The purpose of the present invention is to provide a model predictive control system and method for a medicine calcining machine based on the entropy weight method to solve the problems of inaccurate temperature control, slow response, and high energy consumption when using a medicine calcining machine to process medicinal materials, and improve the control accuracy and efficiency of the system.

[0005] To achieve the above tasks, the present invention adopts the following technical solutions:

[0006] A model predictive control system for a medicine calcining machine based on the entropy weight method, comprising:

[0007] A medicinal material identification module, configured to obtain the smell, shape, and color data of the medicinal material after the medicinal material enters the medicine calcining machine;

[0008] A processing information database, including two sub-databases; wherein, the first sub-database pre-stores information entries of medicinal materials for realizing the matching of the processing time and processing temperature of the medicinal materials; the second sub-database pre-stores process entries of medicinal materials for realizing the matching of the initial power of the electric heating machine;

[0009] A real-time monitoring module, including three sub-monitoring modules; the first sub-monitoring module is used to collect quality data during the processing of the medicinal material, the second sub-monitoring module is used to collect temperature data at different positions during the processing of the medicinal material, and the third sub-monitoring module is used to collect the real-time power of the electric heating machine;

[0010] The model prediction module, based on the medicinal material processing information, the quality data, temperature data obtained in real time, and the real-time power of the electrothermal machine, uses the model predictive control method of the calcining machine based on the entropy weight method to predict and output the power of the electrothermal machine in the calcining machine at the next moment.

[0011] The temperature control module is used to control the heating module of the electrothermal machine to heat and work at the power output at the next moment.

[0012] Furthermore, the medicinal material identification module includes an electronic nose for identifying the odor of the medicinal material and a high-definition camera for identifying the shape and color of the medicinal material; when the medicinal material enters the calcining machine, the odor sampling operator and gas sensor array in the electronic nose identify the odor intensity and odor category of the medicinal material, convert them into digital signals of the odor, and transmit them together with the shape and color obtained and identified by the high-definition camera to the processing information database.

[0013] Furthermore, each information entry in the first sub-database includes the type of medicinal material, the name of the medicinal material, shape, color, odor, processing time, and processing temperature; using the shape, odor, and color collected by the medicinal material identification module as retrieval conditions, the first sub-database matches the stored information entries with the retrieval conditions, finds the corresponding type of medicinal material and the name of the medicinal material, and obtains the processing time and processing temperature of the medicinal material from the corresponding information entry.

[0014] Each process entry in the second sub-database includes the type of medicinal material and the initial power of the electrothermal machine matching the medicinal material of this type; the second sub-database selects the initial power of the electrothermal machine in the corresponding process entry according to the type of medicinal material matched by the first sub-database, and transmits the initial power of the electrothermal machine, processing time, and processing temperature as medicinal material processing information to the real-time monitoring module.

[0015] Furthermore, the first sub-monitoring module includes multiple quality sensors for detecting the real-time quality of the medicinal material placed on the heating medium; the second sub-monitoring module includes multiple temperature sensors for monitoring the real-time temperature at various positions of the calcining machine during the medicinal material processing, including the initial temperature of the air in the calcining machine, the real-time temperature of the heating medium, the real-time temperature of the medicinal material, the real-time temperature of the air in the calcining machine, and the real-time temperature of the machine shell; the third sub-monitoring module is used to collect the real-time power of the electrothermal machine, and transmit these quality data, temperature data, the real-time power of the electrothermal machine, and the medicinal material processing information to the model prediction module.

[0016] A model predictive control method for a calcining machine based on the entropy weight method includes:

[0017] Construct a non-linear model based on the quality data, temperature data, and real-time power of the electrothermal machine during the medicinal material processing; perform a conversion process on the non-linear model to construct a system state space equation, and determine the coefficient matrix of the system state space equation.

[0018] For the established system state - space equation, the backstepping method is applied to linearize it, and a state observer is designed to observe the value of the first - order derivative of the real - time temperature of the medicinal materials;

[0019] The zero - order hold is used to discretize the system state - space equation obtained by linearization, and a discrete - time system is constructed;

[0020] For the discrete - time system, the rolling - optimization method is used to predict the system state for a future period of time;

[0021] A quadratic control law is designed. The entropy - weight method is used to obtain the weight coefficients of the system state and the input variables in the quadratic control law; the predicted system state for a future period of time is substituted, and constraint conditions are set to solve the sequence of system input variables. The first input variable in the solved sequence is used as the power of the electro - thermal machine at the next moment and applied to the medicine - forging machine.

[0022] Furthermore, the non - linear model is as follows:

[0023] ;

[0024] ;

[0025] Where, is the real - time power of the electro - thermal machine, is the working time of the electro - thermal machine; is the initial temperature of the air in the medicine - forging machine; is the mass of the heating medium, is the specific heat capacity of the heating medium, is the real - time temperature of the heating medium; is the real - time mass of the medicinal materials, is the specific heat capacity of the medicinal materials, is the real - time temperature of the medicinal materials; is the real - time temperature of the air in the medicine - forging machine; is the real - time temperature of the medicine - forging machine housing; the parameter superscripts 、 represent its first - order and second - order derivatives; 、 、 、 are all composite functions;

[0026] Let the state variable where and the superscript T represents transpose; the input variable u = p, and the output variable y is the real - time temperature T2 of the medicinal materials. The system state - space equation is established from the above non - linear model:

[0027] ;

[0028] ;

[0029] wherein, is the first derivative of the system state variable, u is the input variable, and y is the output variable; A is the system matrix, B is the control matrix, C is the output matrix or observation matrix, and D is the direct transmission matrix;

[0030] ;

[0031] .

[0032] Furthermore, for the established system state space equation, backstepping is applied to linearize it, and a state observer is designed to observe the value of the first derivative of the real-time temperature of the medicinal material, including:

[0033] Using backstepping, we can obtain:

[0034] ;

[0035] where T m is the target temperature during the processing of the medicinal material, i.e., the processing temperature; is an adjustable parameter, , are the first and second derivatives of the real-time temperature T2 of the medicinal material, then there is:

[0036] ;

[0037] Therefore, the linearized system state space equation can be expressed as:

[0038] ;

[0039] ;

[0040] Design a state observer to observe Let the state variable of the state observer = , , where ; The first derivative of = , , and the result of the state observer is:

[0041] ;

[0042] ;

[0043] wherein, , is the state variable of the state observer , is the estimated value of , is the estimated value of the first derivative of the state variable in the state observer is the observer gain matrix;

[0044] Let the error e between the first derivative of the state variable x in the system state - space equation and the first derivative of the state variable in the state observer tend to 0, and calculate the state variable x of the system state - space equation

[0045] Furthermore, the zero - order hold is used to discretize the system state - space equation obtained by linearization, and a discrete system is constructed, expressed as:

[0046] ;

[0047] where k is the current time is the value of the state variable x at time k is the value of the input variable u at time k represents the predicted system state at time k + 1 at time k , are the discrete - system coefficient matrices, and there are:

[0048] ;

[0049] ;

[0050] where I is the identity matrix, A is the system matrix, B is the control matrix, and T is the sampling period

[0051] Furthermore, the design of the quadratic - form control law includes:

[0052] Design the general form of the quadratic - form control law J:

[0053] ;

[0054] where Q is the weight coefficient of the system state at time , R is the weight coefficient of the input variable, and F is the weight coefficient of the system state at time is the predicted system state at time k + i at time k,

[0055] the constraint condition of the quadratic - form control law J is:

[0056] 。

[0057] Furthermore, the entropy weight method is used to obtain the weight coefficients of the system state and the input variables in the quadratic control law, including:

[0058] Obtain the historical data of the currently processed medicinal materials, and extract the real-time temperature T2 of the medicinal materials, the derivative of the real-time temperature of the medicinal materials and the power p at the k+1 to k+N moments in the same prediction interval length. Each data value of each type of historical data is used as a sample;

[0059] Establish a positive index for each sample :

[0060] ;

[0061] where is the positive index of sample a ij , min and max are the operations of finding the minimum and maximum values, and a ij represents the data value at the k+i moment of the jth type of historical data;

[0062] Then, calculate the proportion of the positive index of each sample in the positive index at all moments , the entropy value of the positive index corresponding to each type of historical data :

[0063] ;

[0064] ;

[0065] where ;

[0066] Finally, obtain the weight of the positive index of each type of historical data :

[0067] ;

[0068] where ; The weight coefficient Q of the system state at the moment is a positive definite matrix, and each element on its diagonal is composed of ; The weight coefficient R is an identity matrix, and its diagonal elements are ;

[0069] Then the weight coefficient F of the system state at the moment is:

[0070] 。

[0071] Compared with the prior art, the present invention has the following technical features:

[0072] 1. The present invention integrates modules such as medicinal material identification, real-time monitoring, and model prediction, maps the data of the power control and temperature control systems, and realizes the coupled control of power and temperature by constructing a non-linear model; adopts the model predictive control method to improve the robustness and stability of the temperature control system and realizes the dynamic and precise control of temperature.

[0073] 2. The present invention applies large model technology to compare the collected medicinal material trait data with the historical data in the database, and uses an electronic nose and a high-definition camera to replace manual identification and manual selection, reducing labor costs and working hours, and realizing the automation of the medicinal material identification work.

[0074] 3. The present invention dynamically updates the weight coefficients of the linear quadratic control law, applies entropy weight to obtain the weight coefficients of the system state and system input; performs the same processing on the real-time data in the database, continuously updates the weight coefficients, makes the prediction result closer to the expected trajectory, and realizes more accurate and rapid temperature control. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 is the structural diagram of the system of the present invention;

[0076] Figure 2 is the schematic diagram of the model establishment and data analysis process in an embodiment of the method of the present invention;

[0077] Figure 3 is the schematic diagram of the data processing process in an embodiment of the method of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0078] Referring to the attached Figure 1 , a model predictive control system of a medicinal material calcining machine based on the entropy weight method provided by the present invention includes:

[0079] A medicinal material identification module, which includes an electronic nose for identifying the smell of medicinal materials and a high-definition camera for identifying the shape and color of medicinal materials; when the medicinal materials enter the calcining machine, the smell sampling operator and gas sensor array in the electronic nose identify the smell intensity and smell category of the medicinal materials, convert them into digital signals of the smell, and transmit them to the processing information database together with the shape and color obtained and identified by the high-definition camera.

[0080] Processing information database, which contains two sub-databases; among them, the first sub-database pre-stores information entries of medicinal materials, and each information entry includes: types of medicinal materials, names of medicinal materials, shapes, colors, odors, processing time, processing temperature, etc.; using the shape, odor, and color collected by the medicinal material recognition module as retrieval conditions, the first sub-database matches the stored information entries with the retrieval conditions, finds the corresponding types of medicinal materials and names of medicinal materials, and obtains the processing time t of the medicinal materials from the corresponding information entries g and the processing temperature T m .

[0081] The second sub-database pre-stores process entries of medicinal materials, and each process entry includes: types of medicinal materials and the initial power of the electric heating machine matching the medicinal materials of this type ; The second sub-database selects the initial power of the electric heating machine in the corresponding process entry according to the type of medicinal materials matched by the first sub-database , and takes the initial power of the electric heating machine , processing time t g , processing temperature T m as the medicinal material processing information and transmits it to the real-time monitoring module.

[0082] Real-time monitoring module, which contains three sub-monitoring modules, and is used to collect quality data, temperature data, and real-time power of the electric heating machine during the processing of medicinal materials; the first sub-monitoring module includes multiple quality sensors, which are used to detect the real-time quality of the medicinal materials placed on the heating medium ; The second sub-monitoring module includes multiple temperature sensors, which are installed at multiple positions such as the heating medium and the casing of the medicine calcining machine, and are used to monitor the real-time temperature of each position of the medicine calcining machine during the processing of medicinal materials. The real-time temperature to be monitored includes the initial temperature T0 of the air in the medicine calcining machine, the real-time temperature T1 of the heating medium, the real-time temperature T2 of the medicinal materials, the real-time temperature T3 of the air in the medicine calcining machine, the real-time temperature T4 of the casing, etc.; the third sub-monitoring module is used to collect the real-time power p of the electric heating machine;

[0083] and transmits these quality data, temperature data, real-time power of the electric heating machine, and medicinal material processing information to the model prediction module.

[0084] Model prediction module, based on the medicinal material processing information, real-time obtained quality data, temperature data, and real-time power of the electric heating machine, uses the model prediction control method of the medicine calcining machine based on the entropy weight method to predict and output the power of the electric heating machine in the medicine calcining machine at the next moment;

[0085] Temperature control module, which is used to control the heating module of the electric heating machine to heat and work with the power output at the next moment.

[0086] In an embodiment of the present invention, taking the processing of keel as an example, an electronic nose is installed outside the feeding port of the medicine calcining machine to identify the smell of the keel; a high-definition camera is arranged on the base of the electronic nose to identify the shape and color of the keel; a mass sensor is installed at the bottom of the crucible of the medicine calcining machine to monitor the change in the quality of the medicinal materials; temperature sensors are evenly arranged at equal intervals at the bottom of the crucible, the top of the crucible, inside the casing and inside the medicine calcining machine, and 2 temperature sensors are arranged at each monitoring point to monitor the temperature change at different positions.

[0087] In this solution, the model prediction control method of the medicine calcining machine based on the entropy weight method in the model prediction module is as follows:

[0088] Step 1: Construct a nonlinear model based on the quality data, temperature data and real-time power of the electric heating machine during the medicinal material processing process; perform transformation processing on the nonlinear model to construct a system state space equation and determine the coefficient matrix of the system state space equation.

[0089] This step uses the heating and heat conduction mechanism for system identification, constructs a nonlinear model, and performs transformation processing on the nonlinear model to obtain a state space equation; first, system identification is required. Since the model of the medicine calcining machine has less heat dissipation and a simple structure, the following nonlinear model is established:

[0090] ;

[0091] ;

[0092] Among them, is the real-time power of the electric heating machine, is the working time of the electric heating machine; is the initial temperature of the air in the medicine calcining machine; is the mass of the heating medium (in this solution, a heating plate is used for heating, so it is a constant value), is the specific heat capacity of the heating medium, is the real-time temperature of the heating medium; is the real-time mass of the medicinal materials, is the specific heat capacity of the medicinal materials, is the real-time temperature of the medicinal materials; is the real-time temperature of the air in the medicine calcining machine; is the real-time temperature of the casing of the medicine calcining machine.

[0093] In this solution, the parameter superscripts 、 represent their first-order and second-order derivatives. For example, is the first-order derivative of the real-time temperature of the heating medium, is the first-order derivative of the real-time temperature of the medicinal materials, the first-order derivative of the real-time temperature of the air inside the medicine calcining machine, is the first derivative of the real-time temperature of the medicine forging machine housing; 、 、 、 are all composite functions.

[0094] Let the state variable , where ; the input variable u = p, and the output variable y is the real-time temperature T2 of the medicinal material. The system state space equation is established from the above nonlinear model:

[0095] ;

[0096] ;

[0097] where, is the first derivative of the system state variable, u is the input variable, and y is the output variable; the superscript indicates that T is the transpose.

[0098] According to 、 、 、 's coefficients and the general form of the state space equation, determine the coefficient matrix of the system state space equation:

[0099] where, A is the system matrix, B is the control matrix, C is the output matrix or observation matrix, and D is the direct transfer matrix:

[0100] ;

[0101] ;

[0102] where, is the second derivative of , is the second derivative of .

[0103] Step 2, for the established system state space equation, apply the backstepping method to linearize it and design a state observer to observe the value of the first derivative of the real-time temperature of the medicinal material.

[0104] To make the system state space equation asymptotically stable, a Lyapunov function needs to be obtained, and using the backstepping method, we can get:

[0105] ;

[0106] where, T m is the target temperature during the processing of the medicinal material, that is, the processing temperature; is an adjustable parameter, 、 are the first and second derivatives of the real-time temperature T2 of the medicinal material, then there are:

[0107] ;

[0108] Therefore, the linearized system state space equation can be expressed as:

[0109] ;

[0110] ;

[0111] By designing the above backstepping method, the state space equation is linearized.

[0112] In the above formula, the specific numerical values and performances cannot be directly observed, which leads to the inability to calculate the results of the above state space equation. It is necessary to design a state observer to observe .

[0113] Let the state variables of the state observer = , , where ; the output variable is ; the above is the disturbance term, which will cancel out with the input or can be regarded as the initial state. Then there is the initial state space equation applied to the state observer:

[0114] ;

[0115] ;

[0116] Denote 's first derivative as = , .

[0117] Through the design and calculation of the above state observer, the final observer result can be obtained as:

[0118] ;

[0119] ;

[0120] Among them, , are the estimated values of the state variables , of the state observer, , are the estimated values of the first derivatives of the state variables in the state observer, is the observer gain matrix.

[0121] Let the first derivative of the state variable x of the system state - space equation and the error e between the first derivative of the state variable of the state observer tend to 0, and calculate the state variable x of the system state - space equation, so as to obtain , for subsequent calculation processes:

[0122] .

[0123] Step 3: Use a zero - order hold to discretize the system state - space equation obtained by linearization processing, and construct a discrete system.

[0124] Since the system state - space equation obtained in Step 2 is continuous, to simplify the model, improve the model stability, reduce the risk of overfitting, and improve the calculation efficiency, apply a zero - order hold to discretize the state - space equation after linearization processing to obtain a discrete system:

[0125] ;

[0126] where k is the current time,[[]] is the value of the state variable x at time k,[[]] is the value of the input variable u at time k,[[]] represents the predicted system state at time k + 1 at time k,[[]] , are the coefficient matrices of the discrete system, and there are:

[0127] ;

[0128] ;

[0129] where I is the identity matrix,[[]] is the time constant,[[]] is the natural constant, A is the system matrix, B is the control matrix, and T is the sampling period.[[]]

[0130] Step 4: For the discrete system, use the rolling - optimization method to predict the system state for a period of time in the future.[[]]

[0131] The system state for a period of time in the future to has the following expression:

[0132] ; ; ;... ;

[0133] where,[[]] is the initial system state at time k, and its value is the state variable x solved in Step 2; is the predicted system state at time k for time k+N; is the input variable at time k, and its value is the real-time power p of the electrothermal machine, is the predicted input variable at time k for time k+N-1, where N is the prediction interval length.

[0134] Step 5: Design the quadratic control law, and obtain the weight coefficients of the system state and the weight coefficients of the input variable in the quadratic control law by using the entropy weight method; Substitute the predicted system state in the future period of time, and set the constraint conditions to solve the system input variable sequence, and use the first input variable in the solved sequence as the power of the electrothermal machine at the next moment and apply it to the medicine forging machine.

[0135] First, design the general form of the quadratic control law J:

[0136] ;

[0137] where Q is the weight coefficient of the system state at time , R is the weight coefficient of the input variable, and F is the weight coefficient of the system state at time is the predicted system state at time k for time k+i obtained through Step 4,

[0138] The temperature fluctuation range of the medicine forging machine is usually designed within ±1°C to ensure the stability and accuracy of the medicine forging process; for the above quadratic control law J, its constraint conditions are:

[0139] ;

[0140] For the above quadratic control law J, its weight coefficient Q of the system state at time weight coefficient F of the system state at time, and the weight coefficient R of the input variable are obtained by using the entropy weight method, as follows:

[0141] Obtain the historical data of the currently processed medicinal materials, and extract the real-time temperature T2 of the medicinal materials, the derivative of the real-time temperature of the medicinal materials and the power p of these three historical data at times k+1 to k+N in the same prediction interval length, and use each data value of each historical data as a sample; Represent it in the form of a table with N rows and 3 columns:

[0142]

[0143] where the sample \(a\) in the \(i\) -th (\(i = 1,2,\cdots,N\)) row and \(j\) -th (\(j = 1,2,3\)) column ij is the data value at the \((k + i)\) -th moment of the \(j\) -th historical data; a positive index is established for each sample :

[0144] ;

[0145] where is the positive index of the sample \(a\) ij , and \(min\), \(max\) are the operations of finding the minimum value and the maximum value.

[0146] Then, calculate the proportion of the positive index of each sample in the positive indices at all moments , and the entropy value of the positive index corresponding to each type of historical data :

[0147] ;

[0148] ;

[0149] where ;

[0150] Finally, obtain the weight of the positive index of each type of historical data :

[0151] ;

[0152] where the weight coefficient \(Q\) of the system state at the moment is a positive definite matrix, and each element on its diagonal is composed of ; the weight coefficient \(R\) is an identity matrix, and its diagonal elements are .

[0153] Then the weight coefficient \(F\) of the system state at the moment is:

[0154] ;

[0155] Take the first \(N - 1\) values in the system state in the future period obtained in step 4, that is to as , and take the \(N\) -th value as Substitute it into the quadratic - form control law \(J\), take the derivative of the quadratic - form control law \(J\), and set the derivative to 0, then the input variable sequence of the system , \(i = 1\cdots N - 1\).

[0156] The obtained system input variable is the combined matrix of the input variables u of the system from the k-th moment to the k+N-1-th moment, and the first input variable is taken. It is applied to the medicine forging machine as the power p of the electrothermal machine at the next moment, and the electrothermal machine is controlled to generate heat at this power.

[0157] When the present invention is specifically applied, the working process is as follows:

[0158] At the initial moment k, the medicine forging machine first starts to work with the initial power and sets the processing time t g , the processing temperature T m , and starts the first cycle. The process of each cycle is as follows:

[0159] The real-time monitoring module obtains the medicine temperature T2 at this time as , the derivative of T2 is ; Since the specific value and performance of cannot be directly observed, the state observer in step 2 is used to solve and obtain , so as to obtain the state variables of the state space equation . This state variable x is used as the initial state of the system at the k-th moment in the rolling optimization method in step 4 , and the system state at future moments is predicted and obtained to . Then to are used as and substituted into the quadratic control law J in step 5. By taking the derivative of J and setting the derivative equal to zero, the system input variable sequence is obtained, and its first value is used as the power p of the electrothermal machine at the next moment and applied to the medicine forging machine. The temperature control module controls the heating module of the electrothermal machine to generate heat at this power.

[0160] At the next moment k+1, the medicine forging machine starts to work with the new power p, starts the second cycle; and repeats the above cycle process until the medicine processing is completed or both the set processing time t g and the processing temperature T m are reached.

[0161] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A model predictive control system for a medicine calcining machine based on the entropy weight method, characterized in that, Including: A medicinal material identification module, which is used to obtain the odor, shape, and color data of the medicinal material after the medicinal material enters the medicine forging machine; A processing information database, which includes two sub-databases; among them, the first sub-database pre-stores information entries of medicinal materials for matching the processing time and processing temperature of medicinal materials; the second sub-database pre-stores process entries of medicinal materials for matching the initial power of the electric heating machine; A real-time monitoring module, which includes three sub-monitoring modules; the first sub-monitoring module is used to collect quality data during the processing of medicinal materials, the second sub-monitoring module is used to collect temperature data at different positions during the processing of medicinal materials, and the third sub-monitoring module is used to collect the real-time power of the electric heating machine; A model prediction module, based on the medicinal material processing information, the quality data, temperature data, and real-time power of the electric heating machine obtained in real time, uses the model predictive control method of the medicine forging machine based on the entropy weight method to predict and output the power of the electric heating machine in the medicine forging machine at the next moment; the method includes: Constructing a nonlinear model based on the quality data, temperature data, and real-time power of the electric heating machine during the processing of medicinal materials; performing transformation processing on the nonlinear model, constructing a system state space equation, and determining the coefficient matrix of the system state space equation; For the established system state space equation, applying the backstepping method to linearize it, and designing a state observer to observe the value of the first derivative of the real-time temperature of the medicinal material; Using a zero-order hold to discretize the system state space equation obtained by the linearization process to construct a discrete system; For the discrete system, using the rolling optimization method to predict the system state for a period of time in the future; Designing a quadratic control law, using the entropy weight method to obtain the weight coefficients of the system state and the weight coefficients of the input variables in the quadratic control law; substituting the predicted system state for a period of time in the future, and setting constraint conditions, solving the system input variable sequence, and taking the first input variable in the solved sequence as the power of the electric heating machine at the next moment and applying it to the medicine forging machine; A temperature control module, which is used to control the heating module of the electric heating machine to heat and work with the power output at the next moment.

2. The predictive control system of the medicine calcining machine based on the entropy weight method according to claim 1, wherein The medicinal material identification module includes an electronic nose for identifying the odor of the medicinal material and a high-definition camera for identifying the shape and color of the medicinal material; when the medicinal material enters the medicine forging machine, the odor sampling operator and gas sensor array in the electronic nose identify the odor intensity and odor category of the medicinal material, convert it into a digital signal of the odor, and transmit it to the processing information database together with the shape and color obtained and identified by the high-definition camera.

3. The predictive control system of the medicine calcining machine based on the entropy weight method according to claim 2, wherein Each information entry in the first sub-database includes the type of medicinal material, name of the medicinal material, shape, color, odor, processing time, and processing temperature; using the shape, odor, and color collected by the medicinal material identification module as retrieval conditions, the first sub-database matches the stored information entries with the retrieval conditions, finds the corresponding type and name of the medicinal material, and obtains the processing time and processing temperature of the medicinal material from the corresponding information entry; Each process entry in the second sub-database includes the type of medicinal material and the initial power of the electrothermal machine matching the medicinal material of this type; the second sub-database selects the initial power of the electrothermal machine in the corresponding process entry according to the type of medicinal material matched by the first sub-database, and transmits the initial power of the electrothermal machine, the processing time, and the processing temperature as medicinal material processing information to the real-time monitoring module.

4. The predictive control system of the medicine calcining machine based on the entropy weight method according to claim 3, wherein The first sub-monitoring module includes a plurality of quality sensors for detecting the real-time quality of the medicinal materials placed on the heating medium; the second sub-monitoring module includes a plurality of temperature sensors for monitoring the real-time temperature at various positions of the medicine calcining machine during the medicinal material processing, including the initial temperature of the air in the medicine calcining machine, the real-time temperature of the heating medium, the real-time temperature of the medicinal materials, the real-time temperature of the air in the medicine calcining machine, and the real-time temperature of the machine shell; the third sub-monitoring module is used to collect the real-time power of the electrothermal machine, and transmit these quality data, temperature data, the real-time power of the electrothermal machine, and the medicinal material processing information to the model prediction module.

5. The predictive control system of the medicine calcining machine based on the entropy weight method according to claim 1, wherein The non-linear model is as follows: ； ； Among them, is the real-time power of the electrothermal machine, is the working time of the electrothermal machine; is the initial temperature of the air in the medicine calcining machine; is the mass of the heating medium, is the specific heat capacity of the heating medium, is the real-time temperature of the heating medium; is the real-time mass of the medicinal materials, is the specific heat capacity of the medicinal materials, is the real-time temperature of the medicinal materials; is the real-time temperature of the air in the medicine calcining machine; is the real-time temperature of the casing of the medicine calcining machine; The parameter superscript , represent its first-order and second-order derivatives; , , , are all composite functions; Let the state variable , where , the superscript T represents transpose; the input variable u = p, and the output variable y is the real-time temperature T2 of the medicinal material. The system state-space equation is established from the above nonlinear model: ； ； wherein, is the first derivative of the system state variable, u is the input variable, and y is the output variable; A is the system matrix, B is the control matrix, C is the output matrix or observation matrix, and D is the direct transfer matrix; ； 。 6. The predictive control system of the medicine calcining machine based on the entropy weight method according to claim 5, wherein For the established system state space equation, backstepping is applied to linearize it, and a state observer is designed to observe the value of the first derivative of the real-time temperature of the medicinal materials, including: Using backstepping, we can obtain: ； Among them, T m is the target temperature during the processing of medicinal materials, that is, the processing temperature; is an adjustable parameter, , are the first and second derivatives of the real-time temperature T2 of the medicinal materials, respectively, then there is: ； Therefore, the linearized system state space equation can be expressed as: ； ； A design state observer is used to observe , and let the state variables of the state observer = , , where ; The first derivative of = , , and the result of the state observer is: ； ； Among them, and are the estimated values of the state observer state variables and . and are the estimated values of the first-order derivatives of the state variables in the state observer, is the observer gain matrix; Let the first derivative of the state variable x of the system state space equation The error e between the first derivative of the state variable of the state observer approaches 0, and the state variable x of the system state space equation is solved.

7. The prediction control system of the medicine calcining machine based on the entropy weight method according to claim 5, characterized in that The system state space equation obtained by linearization is discretized using a zero-order hold to construct a discrete system, expressed as: ； where k is the current time, is the value of the state variable x at time k, is the value of the input variable u at time k, represents the predicted system state at time k+1 at time k, and is the discrete system coefficient matrix, and we have: ； ； Where, I is the identity matrix, A is the system matrix, B is the control matrix, and T is the sampling period.

8. The prediction control system of the medicine calcining machine based on the entropy weight method according to claim 7, characterized in that The design of the quadratic control law includes: The general form of the quadratic control law J is designed: ； where Q is the weight coefficient of the system state at a moment, R is the weight coefficient of the input variable, and F is the weight coefficient of the system state at a moment; is the predicted system state at the (k + i)-th moment at the k-th moment, is the predicted input variable at the (k + i)-th moment at the k-th moment, where i = 1, 2, …, N - 1; The constraint conditions of the quadratic control law J are: 。 9. The prediction control system of the medicine calcining machine based on the entropy weight method according to claim 8, wherein, The weight coefficients of the system state and the weight coefficients of the input variables in the quadratic control law are obtained using the entropy weight method, including: Obtain the historical data of the medicinal materials being processed currently, and extract the real-time temperature T2 of the medicinal materials at the k+1 to k+N moments, the derivative of the real-time temperature of the medicinal materials and the three historical data of the power p. Take each data value of each historical data as a sample; Establish positive indicators for each sample : ； Among them, is the positive index of sample a ij , min and max are operations to find the minimum and maximum values, and a ij represents the data value at the k + i-th moment of the j-th historical data; Then, calculate the positive indicators of each sample in turn The proportion in the positive indicators at all times , the entropy value of the positive indicators corresponding to each type of historical data : ； ； Among them, ; Finally, obtain the weights of the positive indicators for each type of historical data : ； Among them, ; The weight coefficient Q of the system state at each moment is a positive definite matrix, and each element on its diagonal is composed of ; The weight coefficient R is an identity matrix, and its diagonal elements are ; Then The weight coefficient F of the system state at a moment is as follows: 。

Images