Conductivity control method based on system identification of water and fertilizer mixing

Abstract

The invention discloses a conductivity control method based on system identification of water and fertilizer mixing, and belongs to the technical field of water and fertilizer all-in-one machines, and the method comprises the following steps: collecting flow and conductivity information of a water and fertilizer mixing system; adjusting an input flow value of a fertilizer through a particle swarm algorithm to obtain a fertilizer liquid of a target EC, and carrying out water and fertilizer mixing system identification and stock solution EC prediction; water and fertilizer premixing and fuzzy control are carried out by determining the estimated suction value of the fertilizer stock solution under the expected EC and fuzzy PI control, the problems that the conductivity value of the fertilizer stock solution cannot be determined and the expected EC value cannot be reached through rapid mixing are solved, the water and fertilizer mixing time is greatly shortened, the EC overshoot phenomenon is avoided, hardware facilities do not need to be added, and the cost is reduced. Original conductivity values of a system mixing model and a complex raw material formula can be rapidly obtained based on model identification; through a premixing-fuzzy PI control strategy, the output fertilizer and pesticide concentration can be quickly and accurately controlled, and a target conductivity value is achieved.

Description

Technical Field

[0001] The invention relates to the technical field of integrated water and fertilizer machines, and in particular to a conductivity control method based on system identification of water and fertilizer mixing. Background Art

[0002] The water-fertilizer mixed conductivity control technology achieves precise fertilizer solution concentration control by monitoring and adjusting the conductivity (EC) of the irrigation solution. Conductivity is an important indicator for measuring the ion concentration in the solution. In the water-fertilizer integrated system, different crops or the same crop at different growth stages require different concentrations of nutrients. Conductivity control can help accurately adjust the water-fertilizer ratio to avoid excessive nutrients causing root damage or insufficient nutrients affecting growth.

[0003] However, in actual applications, delays in the water-fertilizer mixing and transmission process may lead to a lag in conductivity response, affecting the control accuracy and speed. How to quickly and accurately control the water-fertilizer mixing and achieve the target conductivity control has become a technical problem that needs to be solved urgently. Therefore, it is necessary to design a conductivity control method based on system identification of water-fertilizer mixing. Summary of the invention

[0004] The object of the present invention is to provide a conductivity control method for water-fertilizer mixing based on system identification to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a conductivity control method based on system identification of water-fertilizer mixing, comprising the following steps:

[0006] S1. Collect the flow and conductivity information of the water-fertilizer mixing system to obtain the flow of the fertilizer stock solution, the amount of water inhaled, the output fertilizer solution flow and EC value;

[0007] S2. Through the particle swarm algorithm, the input flow value of the fertilizer is adjusted to obtain the fertilizer solution with the target EC, and the water-fertilizer mixing system is identified and the EC of the original solution is predicted;

[0008] S3. Pre-mixing and fuzzy control of water and fertilizer are performed by clarifying the estimated suction value of the fertilizer stock solution under the expected EC and fuzzy PI control.

[0009] In a further embodiment, in step S1, the flow rate of the fertilizer stock solution is counted by the raw material intake flow meter, the output fertilizer liquid flow rate is counted by the output fertilizer liquid flow meter, and the EC value is counted by the output fertilizer liquid EC meter. The system is defined to collect data in the same time period, and the output fertilizer liquid flow rate is recorded as Q(t), the flow rate of the inhaled fertilizer is q(t), the system water intake is P(t), and the output fertilizer liquid EC change is E(t).

[0010] In a further embodiment, in step S2, according to relevant mathematical modeling, the target EC transfer function in the water-fertilizer mixing system is defined as:

[0011]

[0012] Among them, V is the water capacity of the water-fertilizer system, and its value can be obtained by the difference between the water inlet, fertilizer amount and fertilizer output, E I represents the original EC value of the input fertilizer, Q represents the output fertilizer liquid flow value, and τ represents the delay of the water-fertilizer mixing reaction;

[0013] By inputting a fixed flow value of fertilizer, i.e., q(t) = A, the original (1) can be transformed into the following discrete dynamic equation:

[0014]

[0015] Where T is the data sampling time of the acquisition system, u is the system input, which is a fixed value; x is the state variable of the model; and y indicates the output of the model, that is, the EC value of the output fertilizer solution.

[0016] In a further embodiment, in step S2, the collected data is fitted and analyzed. In a fixed water-fertilizer mixing structure, the EC reaction time of the system mixing reaction and the water capacity of the mixing system are usually fixed, that is, τ and V are known, and the main parameters of the mixing model, namely E I , select the particle swarm algorithm for search, the main steps include the following:

[0017] V(i)=ωV(i)+c1r1[P(best,i)-P(i)]+c2r2[P′(global_best)-P(i)] (4)

[0018]

[0019] Where P(i) is a particle of the format (k, θ) at the i-th iteration, V(i) is the velocity of P(i) at the (i+1)-th iteration, is the inertia weight, c1 and c2 represent acceleration constants; r1 and r2 are random values ​​uniformly distributed in [0, 1]; A refers to the parameter E I The function Rand(I) generates a set of parameters E in interval I. I .

[0020] In a further embodiment, in step S2, the global best particle, i.e., P(global_best), will be updated compared with the best particle P(best) in the i-th iteration, wherein the function of particle evaluation is mainly the sum of squared errors (SSE), which is defined as follows:

[0021]

[0022] Where y(t) is the actual collected data, is the predicted value of the mixed model under each particle parameter (k, θ); y and The step length is consistent.

[0023] In a further embodiment, in step S2, after the mixing system identification is completed, the input formula liquid concentration prediction can be completed. When the fertilizer mixing system performs 2-3 mixing cycles, i.e., τ, the model is fitted. When the SSE of the model is less than the value, it indicates that the identification is completed. At this time, E of formula (1) can be obtained. I Value, that is, the EC concentration of the original formula.

[0024] In a further embodiment, in step S3, the original formula EC concentration is E original , given that the system water inflow is P, and the output fertilizer liquid flow is Q, the flow value q' of the corresponding EC fertilizer is determined according to the EC expected value of the fertilizer liquid flow that needs to be output. After determining the flow value q' of the EC fertilizer, a fixed flow input of 2 τ cycles will be performed. During this process, the original fertilizer liquid EC value is synchronously calculated and q' is fine-tuned. If the output fertilizer liquid EC value does not reach 80% of the target value within 2 τ cycles, or the output fertilizer liquid EC value reaches 80% of the target value within 2 τ cycles, the system will intervene in fuzzy PI control to suppress excessive EC excess. The output change of incremental PI is written as follows:

[0025] Δu=u i -u i-1 =K P ΔE+K i e i (8)

[0026] Where K p is the proportional gain, K i is the integral gain; ΔE is the current difference between the i-th measured EC value and the target EC value.

[0027] In a further embodiment, in step S3, in terms of fuzzy definition, a two-input fuzzy controller is adopted, including the difference between the EC error change rate and the target EC value, and the controller defines the fuzzy language value as {NB, NM, NS, Z, PS, PM, PB}, where NB, NM, NS, Z, PS, PM and PB represent negative large, negative medium, negative small, 0, positive small, positive medium and positive large, respectively.

[0028] In a further embodiment, in step S1, during the stable fertilizer mixing process, the water intake and the output fertilizer liquid flow rate change slightly.

[0029] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention solves the problem of being unable to determine the conductivity value of the fertilizer stock solution and quickly mix to reach the expected EC value through the construction of a water-fertilizer mixing system information collection system, mixing system identification and stock solution EC prediction, and pre-mixing and fuzzy control of water and fertilizer, greatly shortens the mixing time of water and fertilizer and avoids EC overshoot, does not require the addition of hardware facilities, and can quickly obtain the system mixing model and the original conductivity value of the complex raw material formula based on model identification; through the pre-mixing-fuzzy PI control strategy, the output fertilizer concentration can be quickly and accurately controlled to reach the target conductivity value. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 is the overall control system diagram of the present invention;

[0031] Figure 2 It is K of water and fertilizer control of the present invention p Fuzzy control rules;

[0032] Figure 3 It is the overall control flow chart of the present invention;

[0033] Figure 4 It is a schematic diagram of the model identification result of the present invention;

[0034] Figure 5 It is a schematic diagram of the model control result of the present invention;

[0035] The accompanying drawings are marked as follows: 1. water source; 2. fertilizer concentrate; 3. raw material intake flow meter; 4. water-fertilizer mixing system; 5. fertilizer liquid output flow meter; 6. fertilizer liquid output EC meter; 7. pipeline. DETAILED DESCRIPTION

[0036] In the following description, a large number of specific details are provided to provide a more thorough understanding of the present invention. However, it is apparent to those skilled in the art that the present invention can be implemented without one or more of these details. In other examples, in order to avoid confusion with the present invention, some technical features well known in the art are not described.

[0037] See also Figure 1-Figure 5 The present invention provides a technical solution: a conductivity control method based on system identification of water-fertilizer mixing, comprising the following steps:

[0038] S1. Collect the flow and conductivity information of the water-fertilizer mixing system to obtain the flow of the fertilizer stock solution, the amount of water inhaled, the output fertilizer solution flow and EC value;

[0039] S2. Through the particle swarm algorithm, the input flow value of the fertilizer is adjusted to obtain the fertilizer solution with the target EC, and the water-fertilizer mixing system is identified and the EC of the original solution is predicted;

[0040] S3. Pre-mixing and fuzzy control of water and fertilizer are performed by clarifying the estimated suction value of the fertilizer stock solution under the expected EC and fuzzy PI control.

[0041] In a further embodiment, in step S1, the flow rate of the fertilizer stock solution is counted by the raw material intake flow meter, the output fertilizer liquid flow rate is counted by the output fertilizer liquid flow meter, and the EC value is counted by the output fertilizer liquid EC meter. The system is defined to collect data in the same time period, and the output fertilizer liquid flow rate is recorded as Q(t), the flow rate of the inhaled fertilizer is q(t), the system water intake is P(t), and the output fertilizer liquid EC change is E(t).

[0042] In a further embodiment, it is known that the water-fertilizer mixing process is a first-order linear delay system, in which the output fertilizer liquid conductivity value (i.e., EC) is related to the input water volume and the original EC value of the fertilizer. By adjusting the input flow value of the fertilizer, the fertilizer liquid of the target EC can be obtained. In step S2, according to the relevant mathematical modeling, the target EC transfer function in the water-fertilizer mixing system is defined as:

[0043]

[0044] Among them, V is the water capacity of the water-fertilizer system, and its value can be obtained by the difference between the water inlet, fertilizer amount and fertilizer output, E I represents the original EC value of the input fertilizer, Q represents the output fertilizer liquid flow value, and τ represents the delay of the water-fertilizer mixing reaction;

[0045] By inputting a fixed flow value of fertilizer, i.e., q(t) = A, the original (1) can be transformed into the following discrete dynamic equation:

[0046]

[0047] Where T is the data sampling time of the acquisition system, u is the system input, which is a fixed value; x is the state variable of the model; and y indicates the output of the model, that is, the EC value of the output fertilizer solution.

[0048] In a further embodiment, in step S2, the collected data is fitted and analyzed. In a fixed water-fertilizer mixing structure, the EC reaction time of the system mixing reaction and the water capacity of the mixing system are usually fixed, that is, τ and V are known, and the main parameters of the mixing model, namely E I , select the particle swarm algorithm for search, the main steps include the following:

[0049] V(i)=ωV(i)+c1r1[P(best,i)-P(i)]+c2r2[P′(global_best)-P(i)] (4)

[0050]

[0051] Where P(i) is a particle of the format (k, θ) at the i-th iteration, V(i) is the velocity of P(i) at the (i+1)-th iteration, is the inertia weight, c1 and c2 represent acceleration constants; r1 and r2 are random values ​​uniformly distributed in [0, 1]; A refers to the parameter E I The function Rand(I) generates a set of parameters E in interval I. I .

[0052] In a further embodiment, in step S2, the global best particle, i.e., P(global_best), will be updated compared with the best particle P(best) in the i-th iteration, wherein the function of particle evaluation is mainly the sum of squared errors (SSE), which is defined as follows:

[0053]

[0054] Where y(t) is the actual collected data, is the predicted value of the mixed model under each particle parameter (k, θ); y and The step length is consistent, such as Figure 4 As shown, the black dotted line is the collected data y(t), and the red line is the estimated value after the identification model

[0055]

[0056] In a further embodiment, in step S2, after the mixing system identification is completed, the input formula liquid concentration prediction can be completed. When the fertilizer mixing system performs 2-3 mixing cycles, i.e., τ, the model is fitted. When the SSE of the model is less than the value, it indicates that the identification is completed. At this time, E of formula (1) can be obtained. I Value, that is, the EC concentration of the original formula. This process can be predicted during the operation. If the formula of the current operation is consistent with that of the historical operation, the above system identification method can be used to determine the model and estimate the EC concentration of the original formula.

[0057] In a further embodiment, in step S3, the original formula EC concentration is E original, given that the system water inflow is P, and the output fertilizer liquid flow is Q, the flow value q' of the corresponding EC fertilizer is determined according to the EC expected value of the fertilizer liquid flow that needs to be output. After determining the flow value q' of the EC fertilizer, a fixed flow input of 2 τ cycles will be performed. During this process, the original fertilizer liquid EC value is synchronously calculated and q' is fine-tuned. If the output fertilizer liquid EC value does not reach 80% of the target value within 2 τ cycles, or the output fertilizer liquid EC value reaches 80% of the target value within 2 τ cycles, the system will intervene in fuzzy PI control to suppress excessive EC excess. The output change of incremental PI is written as follows:

[0058] Δu=u i -u i-1 =K P ΔE+K i e i (8)

[0059] Where K p is the proportional gain, K i is the integral gain; ΔE is the current difference between the i-th measured EC value and the target EC value, such as Figure 5 As shown in FIG. 1 , this process is based on system identification under historical data, that is, the conductivity concentration of the formula is known before startup.

[0060] In a further embodiment, in step S3, in terms of fuzzy definition, a two-input fuzzy controller is adopted, including the difference between the EC error change rate and the target EC value, and the controller defines the fuzzy language value as {NB, NM, NS, Z, PS, PM, PB}, where NB, NM, NS, Z, PS, PM and PB represent negative large, negative medium, negative small, 0, positive small, positive medium and positive large, respectively.

[0061] In a further embodiment, in step S1, during the stable fertilizer mixing process, the water intake and the output fertilizer liquid flow rate change slightly.

[0062] The preferred specific embodiments of the present invention are described in detail above in conjunction with the accompanying drawings. However, the present invention is not limited to the specific details in the above specific embodiments. Within the technical concept of the present invention, various equivalent transformations can be made to the technical scheme of the present invention, and these equivalent transformations all belong to the protection scope of the present invention.

Claims

1. A conductivity control method based on system identification of water-fertilizer mixing, characterized in that: The steps include: S1. Collect the flow and conductivity information of the water-fertilizer mixing system to obtain the flow of the fertilizer stock solution, the amount of water inhaled, the output fertilizer solution flow and EC value; S2. Through the particle swarm algorithm, the input flow value of the fertilizer is adjusted to obtain the fertilizer solution with the target EC, and the water-fertilizer mixing system is identified and the EC of the original solution is predicted; S3. Pre-mixing and fuzzy control of water and fertilizer are performed by clarifying the estimated suction value of the fertilizer stock solution under the expected EC and fuzzy PI control.

2. The conductivity control method based on system identification of water-fertilizer mixing according to claim 1 is characterized in that: In the step S1, the flow rate of the fertilizer stock solution is counted by the raw material intake flow meter, the output fertilizer liquid flow rate is counted by the output fertilizer liquid flow meter, and the EC value is counted by the output fertilizer liquid EC meter. It is defined that the system collects data in the same time period, the output fertilizer liquid flow rate is recorded as Q(t), the intake fertilizer flow rate is q(t), the system water intake is P(t), and the output fertilizer liquid EC change is E(t).

3. A conductivity control method based on system identification of water-fertilizer mixing according to claim 2, characterized in that: In step S2, according to relevant mathematical modeling, the target EC transfer function in the water-fertilizer mixing system is defined as: Among them, V is the water capacity of the water-fertilizer system, and its value can be obtained by the difference between the water inlet, fertilizer amount and fertilizer output, E I represents the original EC value of the input fertilizer, Q represents the output fertilizer liquid flow value, and τ represents the delay of the water-fertilizer mixing reaction; By inputting a fixed flow value of fertilizer, i.e., q(t) = A, the original (1) can be transformed into the following discrete dynamic equation: Where T is the data sampling time of the acquisition system, u is the system input, which is a fixed value; x is the state variable of the model; and y indicates the output of the model, that is, the EC value of the output fertilizer solution.

4. The conductivity control method based on system identification of water-fertilizer mixing according to claim 3 is characterized in that: In step S2, the collected data is fitted and analyzed. In a fixed water-fertilizer mixing structure, the EC reaction time of the system mixing reaction and the water capacity of the mixing system are usually fixed, that is, τ and V are known. In order to determine the main parameters of the mixing model, namely E I , select the particle swarm algorithm for search, the main steps include the following: V(i)=ωV(i)+c1r1[P(best,i)-P(i)]+c2r2[P′(global_best)-P(i)] (4) Where P(i) is a particle of the format (k, θ) at the i-th iteration, V(i) is the velocity of P(i) at the (i+1)-th iteration, is the inertia weight, c1 and c2 represent acceleration constants; r1 and r2 are random values ​​uniformly distributed in [0, 1]; A refers to the parameter E I The function Rand(I) generates a set of parameters E in interval I. I .

5. The conductivity control method based on system identification of water-fertilizer mixing according to claim 4 is characterized in that: In step S2, compared with the best particle P(best) in the i-th iteration, the global best particle, i.e., P(global_best), will be updated, where the function of particle evaluation is mainly the sum of squared errors (SSE), which is defined as follows: Where y(t) is the actual collected data, is the predicted value of the mixed model under each particle parameter (k, θ); y and The step length is consistent.

6. The conductivity control method based on system identification of water-fertilizer mixing according to claim 5 is characterized in that: In step S2, after the mixing system identification is completed, the input formula liquid concentration prediction can be completed. When the mixing fertilizer system performs 2-3 mixing cycles, i.e., τ, the model is fitted. When the SSE of the model is less than the value, it indicates that the identification is completed. At this time, the E of formula (1) can be obtained. I Value, that is, the EC concentration of the original formula.

7. The conductivity control method based on system identification of water-fertilizer mixing according to claim 1 is characterized in that: In step S3, the original formula EC concentration is currently obtained as E original , given that the system water inflow is P, and the output fertilizer liquid flow is Q, the flow value q' of the corresponding EC fertilizer is determined according to the EC expected value of the fertilizer liquid flow that needs to be output. After determining the flow value q' of the EC fertilizer, a fixed flow input of 2 τ cycles will be performed. During this process, the original fertilizer liquid EC value is synchronously calculated and q' is fine-tuned. If the output fertilizer liquid EC value does not reach 80% of the target value within 2 τ cycles, or the output fertilizer liquid EC value reaches 80% of the target value within 2 τ cycles, the system will intervene in fuzzy PI control to suppress excessive EC excess. The output change of incremental PI is written as follows: Δu=u i -u i-1 =K P ΔE+K i e i (8) Where K p is the proportional gain, K i is the integral gain; ΔE is the current difference between the i-th measured EC value and the target EC value.

8. The conductivity control method based on system identification of water-fertilizer mixing according to claim 7 is characterized in that: In step S3, in terms of fuzzy definition, a two-input fuzzy controller is adopted, including the difference between the EC error change rate and the target EC value. The controller defines the fuzzy language value as {NB, NM, NS, Z, PS, PM, PB}, where NB, NM, NS, Z, PS, PM and PB represent negative large, negative medium, negative small, 0, positive small, positive medium and positive large, respectively.

9. The conductivity control method based on system identification of water-fertilizer mixing according to claim 1, characterized in that: In the step S1, during the stable fertilizer mixing process, the water intake and the output fertilizer liquid flow rate change slightly.

Images