A Smart Robust Control Method for Dissolved Oxygen Based on Sliding Mode Observer

Abstract

This invention proposes a robust intelligent control method for dissolved oxygen (DOO) based on a sliding mode observer, achieving accurate control of DDO concentration. The method designs an adaptive predictive fuzzy neural network to predict the dynamic changes in DDO concentration during biochemical reactions, proposes a sliding mode observer to actively suppress disturbances in the biochemical reaction process, and establishes an intelligent robust controller based on the sliding mode observer and the adaptive fuzzy neural network to achieve accurate control of DDO concentration. By designing a predictor to obtain the estimated DDO concentration, and designing an adaptive law based on the prediction error to reduce the impact of prediction error on control performance, the method addresses the problem of low DDO control accuracy. Experimental results show that this method can achieve accurate control of DDO concentration.

Description

Technical Field

[0001] This invention relates to a sliding mode observer-based intelligent robust control method for dissolved oxygen, which enables precise control of dissolved oxygen concentration during wastewater treatment. Dissolved oxygen concentration is one of the key control parameters in the biochemical reaction process of urban wastewater treatment and has a significant impact on the effluent quality. This invention belongs to both the field of water research and the field of intelligent control. Background Technology

[0002] With the acceleration of urbanization and the improvement of people's living standards, the pollution of water resources by various types of domestic waste is becoming increasingly serious. The establishment of wastewater treatment plants is of great significance for improving the urban environment. Currently, the activated sludge process is widely used in urban wastewater treatment. Dissolved oxygen is an important controlled variable in the urban wastewater treatment process, closely related to the biochemical reaction processes of all activated sludge microorganisms. Accurately controlling it within a reasonable range is a prerequisite for the normal operation of the wastewater treatment process. The urban wastewater treatment process is a complex industrial process with strong disturbances and complex nonlinearity, which brings great challenges to the stable control of dissolved oxygen.

[0003] This invention presents a robust intelligent dissolved oxygen control method based on a sliding mode observer. It primarily improves the approximation performance of the neural network for dynamic changes in dissolved oxygen concentration in the fifth zone of the biochemical reaction process by designing an adaptive predictive fuzzy neural network. To mitigate the impact of disturbances during wastewater treatment on dissolved oxygen concentration control, a sliding mode observer is designed to actively suppress disturbances. This invention not only reduces the impact of strong disturbances such as influent flow rate, water quality fluctuations, and changes in operating environment on control performance during wastewater treatment operation, but also approximates the dynamic changes in dissolved oxygen concentration with high accuracy, ensuring that the dissolved oxygen concentration meets the requirements of the biochemical reaction and guaranteeing the safe and stable operation of the wastewater treatment process. Summary of the Invention

[0004] This invention presents a robust intelligent dissolved oxygen control method based on a sliding mode observer. The method employs an adaptive predictive fuzzy neural network to predict the dynamic changes in dissolved oxygen concentration in the fifth zone, while simultaneously utilizing the prediction error to design an adaptive law to update network parameters. The fuzzy neural network approximates the dynamic changes in dissolved oxygen concentration during the biochemical reaction process based on the current dissolved oxygen concentration and the oxygen transfer coefficient, thereby improving control performance to some extent. A sliding mode observer is designed to actively estimate disturbances during the process. The controller controls the dissolved oxygen concentration based on the detected value of the oxygen transfer coefficient, achieving precise control of dissolved oxygen concentration to meet the dissolved oxygen requirements of the biochemical reaction.

[0005] The present invention adopts the following technical solution and implementation steps:

[0006] 1. A method for intelligent robust control of dissolved oxygen concentration based on sliding mode observer, characterized by: establishing a dissolved oxygen concentration prediction model for the biochemical reaction process of urban wastewater treatment based on an adaptive predictive fuzzy neural network, constructing a disturbance information estimation model for the biochemical reaction process of urban wastewater treatment based on a sliding mode observer, and designing an intelligent robust controller based on a sliding mode observer to achieve accurate control of dissolved oxygen concentration; including the following steps:

[0007] (1) Extracting the characteristics of dissolved oxygen concentration changes in the biochemical reaction process of urban sewage treatment

[0008] The dissolved oxygen concentration in the biochemical reaction process of urban wastewater treatment fluctuates significantly over time and is easily affected by changes in influent flow rate, water quality fluctuations, and operating environment. The dissolved oxygen concentration control model for the fifth zone of the biochemical reaction tank is as follows:

[0009]

[0010] Among them, S O,5 (t) represents the dissolved oxygen concentration in the fifth zone of the biochemical reaction tank at time t, h(S) O,5 (t) represents the change in dissolved oxygen concentration in the fifth zone of the biochemical reactor at time t, K L a5(t) represents the oxygen transfer coefficient of the fifth zone of the biochemical reactor at time t, d e (t) represents the external disturbances caused by changes in influent flow, water quality fluctuations, and operating environment during the biochemical reaction process of urban wastewater treatment at time t. d e The dynamic expression for (t) is:

[0011]

[0012] Where Q4(t) is the flow velocity in the fourth zone, and Q4(t) has different values ​​under different weather conditions, S O,4 This represents the dissolved oxygen concentration in the fourth zone of the biochemical reaction tank, with a value of 2 mg / L.

[0013] (2) Establish a prediction model for dissolved oxygen concentration in the biochemical reaction process of urban sewage treatment.

[0014] An adaptive fuzzy neural network prediction model was designed to predict the change in dissolved oxygen concentration h(S) in the fifth zone of the biochemical reaction tank during urban wastewater treatment. O,5 The prediction of (t) is achieved by an adaptive prediction fuzzy neural network with four layers: input layer, membership function layer, rule layer, and output layer. The network structure is 2-ll-1, where l is a positive integer greater than 1 and its initial value is 3. Specifically:

[0015] Input layer: The input layer consists of 2 neurons, g(t) = [g1(t), g2(t)] TLet g1(t) be the input vector of the adaptive prediction fuzzy neural network at time t. O,5 g2(t) = K L a5(t), where T represents the transpose of the vector;

[0016] Membership function layer: The membership function layer has l neurons, using a Gaussian function as the membership function. The output of the membership function layer is:

[0017]

[0018] Where, μ ij (t) represents the output value of the j-th neuron in the membership function layer at time t, i = 1, 2, j = 1, 2, ..., l, g i (t) represents the i-th variable in the input vector g(t) of the input layer, c ij (t) represents the center value of the i-th neuron in the input layer and the j-th neuron in the membership function layer at time t, a ij (t) represents the width values ​​of the i-th neuron in the input layer and the j-th neuron in the membership function layer at time t;

[0019] Rule layer: The rule layer has l neurons, and the output of each neuron in the rule layer is:

[0020]

[0021] Where, γ j (t) represents the output value of the j-th neuron in the regular layer at time t;

[0022] Output layer: The output of the output layer is the predicted value of the change in dissolved oxygen concentration in the fifth partition at time t.

[0023]

[0024] in, Let w(t) be the predicted value of the change in dissolved oxygen concentration in the fifth zone at time t, where w(t) = [w1(t), ..., w l [w(t)] represents the output weight vector of the adaptive prediction fuzzy neural network at time t, and w1(t) represents the output weight of the first neuron in the regular layer of the adaptive prediction fuzzy neural network at time t. l γ(t) represents the output weight of the l-th neuron in the rule layer of the adaptive prediction fuzzy neural network at time t, where γ(t) = [γ1(t), ..., γ2(t)]. l (t)] T Let γt be the output vector of the regular layer neuron in the adaptive prediction fuzzy neural network at time t, and let γ1(t) be the output value of the first neuron in the regular layer of the adaptive prediction fuzzy neural network at time t. l(t) represents the output value of the l-th neuron in the regular layer of the adaptive prediction fuzzy neural network at time t;

[0025] Design a dissolved oxygen concentration predictor for the fifth zone of the urban wastewater treatment biochemical reaction process:

[0026]

[0027] in, S is the derivative of the estimated dissolved oxygen concentration in the fifth partition at time t. O,5d (t) represents the set value of dissolved oxygen concentration at time t, and ε(t) is the prediction error of dissolved oxygen concentration in the fifth zone at time t:

[0028]

[0029] in, This is the estimated value of the dissolved oxygen concentration in the fifth partition at time t;

[0030] Using the dissolved oxygen concentration prediction error ε(t) in the fifth region, an adaptive law for the parameters of an adaptive prediction fuzzy neural network is designed:

[0031] Ξ j (t+1)=Ξ j (t)-0.1[γ j (t)|ε(t)|I+2Ξ j (t)] (8)

[0032] Among them, Ξ j (t)=[w j (t),c j (t),a j [(t)] represents the parameter vector of the j-th neuron in the regular layer of the adaptive prediction fuzzy neural network at time t, w j (t) represents the output weights of the j-th neuron in the rule layer of the adaptive prediction fuzzy neural network at time t, and c j (t) represents the center of the j-th neuron in the regular layer of the adaptive prediction fuzzy neural network at time t, a j (t) represents the width of the j-th neuron in the regular layer of the adaptive prediction fuzzy neural network at time t, I represents a three-dimensional row vector of all 1s, Ξ j (t+1)=[w j (t+1),c j (t+1),a j [(t+1)] represents the parameter vector of the j-th neuron in the regular layer of the adaptive prediction fuzzy neural network at time t+1, w j (t+1) represents the output weight of the j-th neuron in the rule layer of the adaptive prediction fuzzy neural network at time t+1, c ij(t+1) represents the center value of the i-th neuron in the input layer and the j-th neuron in the membership function layer at time t+1, a ij (t+1) represents the width values ​​of the i-th neuron in the input layer and the j-th neuron in the membership function layer at time t+1;

[0033] (3) Design a disturbance information estimation model for the biochemical reaction process of urban wastewater treatment.

[0034] Construct a sliding mode observer-based model for estimating disturbance information in the biochemical reaction process of urban wastewater treatment to estimate external disturbances d caused by influent flow, water quality fluctuations, and changes in the operating environment. e The estimate of (t) is as follows:

[0035]

[0036] in,

[0037]

[0038]

[0039] in, This represents the external perturbation d during the biochemical reaction process obtained through a sliding mode observer. e (t) is the derivative of the observed value. sgn() represents the sign function, and χ1(t) represents the S value during the biochemical reaction process obtained through the sliding mode observer. O,5 The observed value of χ²(t), where χ²(t) represents the first derivative of the external perturbation during the biochemical reaction obtained through the sliding mode observer. The observed values, This indicates that S represents the biochemical reaction process obtained through a sliding mode observer. O,5 (t) is the derivative of the observed value. This represents the external perturbation d during the biochemical reaction process obtained through a sliding mode observer. e (t) The second derivative of the observed value, according to formula (11) for external disturbances d in the biochemical reaction process. e (t) is estimated;

[0040] (4) Design a smart robust control law for dissolved oxygen based on a sliding mode observer.

[0041] ① At the initial time t=1 of the control action, the initial values ​​of the adaptive prediction fuzzy neural network parameters are randomly generated;

[0042] ② Calculate the predicted value of the change in dissolved oxygen concentration in the fifth zone according to formula (5).

[0043] ③ Solve the adaptive law of the adaptive prediction fuzzy neural network parameters according to formula (8);

[0044] ④ Solve for the observed values ​​of influent flow rate, water quality fluctuations, and external disturbances caused by changes in the operating environment during the biochemical reaction process using formula (9). d (t);

[0045] ⑤ Calculate the intelligent robust control law u(t) for dissolved oxygen in the sliding mode observer at time t:

[0046] Calculate the dissolved oxygen concentration tracking and control error in the fifth zone:

[0047] e(t) = S O,5d (t)-S O,5 (t) (12)

[0048] Where e(t) represents the dissolved oxygen concentration setpoint S at time t. O,5d (t) and actual value S O,5 The error of (t);

[0049]

[0050] in, S represents the dissolved oxygen concentration setpoint at time t. O,5d (t) and actual value S O,5 (t) is the derivative of the error, and u(t) is the oxygen transfer coefficient of the controller at time t;

[0051] The intelligent robust control law u(t) for dissolved oxygen at time t is calculated using formulas (5) and (11):

[0052]

[0053] in, This represents the derivative of the dissolved oxygen concentration setpoint at time t.

[0054] ⑥ Increment time t by 1. If t < 200, return to step ②. If t = 200, end the loop.

[0055] (5) Intelligent and robust control of dissolved oxygen concentration

[0056] The dissolved oxygen concentration in the fifth zone of the urban wastewater treatment biochemical reaction process is tracked and controlled using the solved u(t). The output of the control system is the actual dissolved oxygen concentration value. Attached Figure Description

[0057] Figure 1 This is a diagram illustrating the tracking and control effect of the dissolved oxygen fixed setpoint under three weather conditions according to the present invention.

[0058] Figure 2 This is a diagram illustrating the tracking and control effect of the dissolved oxygen setpoint under three weather conditions according to the present invention. Detailed Implementation

[0059] 1. A method for intelligent robust control of dissolved oxygen concentration based on sliding mode observer, characterized by: establishing a dissolved oxygen concentration prediction model for the biochemical reaction process of urban wastewater treatment based on an adaptive predictive fuzzy neural network, constructing a disturbance information estimation model for the biochemical reaction process of urban wastewater treatment based on a sliding mode observer, and designing an intelligent robust controller based on a sliding mode observer to achieve accurate control of dissolved oxygen concentration; including the following steps:

[0060] (1) Extracting the characteristics of dissolved oxygen concentration changes in the biochemical reaction process of urban sewage treatment

[0061] The dissolved oxygen concentration in the biochemical reaction process of urban wastewater treatment fluctuates significantly over time and is easily affected by changes in influent flow rate, water quality fluctuations, and operating environment. The dissolved oxygen concentration control model for the fifth zone of the biochemical reaction tank is as follows:

[0062]

[0063] Among them, S O,5 (t) represents the dissolved oxygen concentration in the fifth zone of the biochemical reaction tank at time t, h(S) O,5 (t) represents the change in dissolved oxygen concentration in the fifth zone of the biochemical reactor at time t, K L a5(t) represents the oxygen transfer coefficient of the fifth zone of the biochemical reactor at time t, d e (t) represents the external disturbances caused by changes in influent flow, water quality fluctuations, and operating environment during the biochemical reaction process of urban wastewater treatment at time t. d e The dynamic expression for (t) is:

[0064]

[0065] Where Q4(t) is the flow velocity in the fourth zone, and Q4(t) has different values ​​under different weather conditions, S O,4 This represents the dissolved oxygen concentration in the fourth zone of the biochemical reaction tank, with a value of 2 mg / L.

[0066] (2) Establish a prediction model for dissolved oxygen concentration in the biochemical reaction process of urban sewage treatment.

[0067] An adaptive fuzzy neural network prediction model was designed to predict the change in dissolved oxygen concentration h(S) in the fifth zone of the biochemical reaction tank during urban wastewater treatment. O,5The prediction of (t) is achieved by an adaptive prediction fuzzy neural network with four layers: input layer, membership function layer, rule layer, and output layer. The network structure is 2-ll-1, where l is a positive integer greater than 1 and its initial value is 3. Specifically:

[0068] Input layer: The input layer consists of 2 neurons, g(t) = [g1(t), g2(t)] T Let g1(t) be the input vector of the adaptive prediction fuzzy neural network at time t. O,5 g2(t) = K L a5(t), where T represents the transpose of the vector;

[0069] Membership function layer: The membership function layer has l neurons, using a Gaussian function as the membership function. The output of the membership function layer is:

[0070]

[0071] Where, μ ij (t) represents the output value of the j-th neuron in the membership function layer at time t, i = 1, 2, j = 1, 2, ..., l, g i (t) represents the i-th variable in the input vector g(t) of the input layer, c ij (t) represents the center value of the i-th neuron in the input layer and the j-th neuron in the membership function layer at time t, a ij (t) represents the width values ​​of the i-th neuron in the input layer and the j-th neuron in the membership function layer at time t;

[0072] Rule layer: The rule layer has l neurons, and the output of each neuron in the rule layer is:

[0073]

[0074] Where, γ j (t) represents the output value of the j-th neuron in the regular layer at time t;

[0075] Output layer: The output of the output layer is the predicted value of the change in dissolved oxygen concentration in the fifth partition at time t.

[0076]

[0077] in, Let w(t) be the predicted value of the change in dissolved oxygen concentration in the fifth zone at time t, where w(t) = [w1(t), ..., w l [w(t)] represents the output weight vector of the adaptive prediction fuzzy neural network at time t, and w1(t) represents the output weight of the first neuron in the regular layer of the adaptive prediction fuzzy neural network at time t. lγ(t) represents the output weight of the l-th neuron in the rule layer of the adaptive prediction fuzzy neural network at time t, where γ(t) = [γ1(t), ..., γ2(t)]. l (t)] T Let γt be the output vector of the regular layer neuron in the adaptive prediction fuzzy neural network at time t, and let γ1(t) be the output value of the first neuron in the regular layer of the adaptive prediction fuzzy neural network at time t. l (t) represents the output value of the l-th neuron in the regular layer of the adaptive prediction fuzzy neural network at time t;

[0078] Design a dissolved oxygen concentration predictor for the fifth zone of the urban wastewater treatment biochemical reaction process:

[0079]

[0080] in, S is the derivative of the estimated dissolved oxygen concentration in the fifth partition at time t. O,5d (t) represents the set value of dissolved oxygen concentration at time t, and ε(t) is the prediction error of dissolved oxygen concentration in the fifth zone at time t:

[0081]

[0082] in, This is the estimated value of the dissolved oxygen concentration in the fifth partition at time t;

[0083] Using the dissolved oxygen concentration prediction error ε(t) in the fifth region, an adaptive law for the parameters of an adaptive prediction fuzzy neural network is designed:

[0084] Ξ j (t+1)=Ξ j (t)-0.1[γ j (t)|ε(t)|I+2Ξ j (t)] (8)

[0085] Among them, Ξ j (t)=[w j (t),c j (t),a j [(t)] represents the parameter vector of the j-th neuron in the regular layer of the adaptive prediction fuzzy neural network at time t, w j (t) represents the output weights of the j-th neuron in the rule layer of the adaptive prediction fuzzy neural network at time t, and c j (t) represents the center of the j-th neuron in the regular layer of the adaptive prediction fuzzy neural network at time t, a j (t) represents the width of the j-th neuron in the regular layer of the adaptive prediction fuzzy neural network at time t, I represents a three-dimensional row vector of all 1s, Ξ j (t+1)=[w j (t+1),cj (t+1),a j [(t+1)] represents the parameter vector of the j-th neuron in the regular layer of the adaptive prediction fuzzy neural network at time t+1, w j (t+1) represents the output weight of the j-th neuron in the rule layer of the adaptive prediction fuzzy neural network at time t+1, c ij (t+1) represents the center value of the i-th neuron in the input layer and the j-th neuron in the membership function layer at time t+1, a ij (t+1) represents the width values ​​of the i-th neuron in the input layer and the j-th neuron in the membership function layer at time t+1;

[0086] (3) Design a disturbance information estimation model for the biochemical reaction process of urban wastewater treatment.

[0087] Construct a sliding mode observer-based model for estimating disturbance information in the biochemical reaction process of urban wastewater treatment to estimate external disturbances d caused by influent flow, water quality fluctuations, and changes in the operating environment. e The estimate of (t) is as follows:

[0088]

[0089] in,

[0090]

[0091] in, This represents the external perturbation d during the biochemical reaction process obtained through a sliding mode observer. e (t) is the derivative of the observed value. sgn() represents the sign function, and χ1(t) represents the S value during the biochemical reaction process obtained through the sliding mode observer. O,5 The observed value of χ²(t), where χ²(t) represents the first derivative of the external perturbation during the biochemical reaction obtained through the sliding mode observer. The observed values, This indicates that S represents the biochemical reaction process obtained through a sliding mode observer. O,5 (t) is the derivative of the observed value. This represents the external perturbation d during the biochemical reaction process obtained through a sliding mode observer. e (t) The second derivative of the observed value, according to formula (11) for external disturbances d in the biochemical reaction process. e (t) is estimated;

[0092] (4) Design a smart robust control law for dissolved oxygen based on a sliding mode observer.

[0093] ① At the initial time t=1 of the control action, the initial values ​​of the adaptive prediction fuzzy neural network parameters are randomly generated;

[0094] ② Calculate the predicted value of the change in dissolved oxygen concentration in the fifth zone according to formula (5).

[0095] ③ Solve the adaptive law of the adaptive prediction fuzzy neural network parameters according to formula (8);

[0096] ④ Solve for the observed values ​​of influent flow rate, water quality fluctuations, and external disturbances caused by changes in the operating environment during the biochemical reaction process using formula (9). d (t);

[0097] ⑤ Calculate the intelligent robust control law u(t) for dissolved oxygen in the sliding mode observer at time t:

[0098] Calculate the dissolved oxygen concentration tracking and control error in the fifth zone:

[0099] e(t) = S O,5d (t)-S O,5 (t) (12)

[0100] Where e(t) represents the dissolved oxygen concentration setpoint S at time t. O,5d (t) and actual value S O,5 The error of (t);

[0101]

[0102] in, S represents the dissolved oxygen concentration setpoint at time t. O,5d (t) and actual value S O,5 (t) is the derivative of the error, and u(t) is the oxygen transfer coefficient of the controller at time t;

[0103] The intelligent robust control law u(t) for dissolved oxygen at time t is calculated using formulas (5) and (11):

[0104]

[0105] in, This represents the derivative of the dissolved oxygen concentration setpoint at time t.

[0106] ⑥ Increment time t by 1. If t < 200, return to step ②. If t = 200, end the loop.

[0107] (5) Intelligent and robust control of dissolved oxygen concentration

[0108] The dissolved oxygen concentration in the fifth zone of the urban wastewater treatment biochemical reaction process is tracked and controlled using the solved u(t). The output of the control system is the actual dissolved oxygen concentration value. Figure 1This is a diagram illustrating the tracking and control effect of the dissolved oxygen fixed setpoint under three weather conditions according to the present invention; Figure 2 This is a graph showing the tracking and control effect of the dissolved oxygen setpoint under three weather conditions. X-axis: time, unit is days; Y-axis: effluent dissolved oxygen concentration and dissolved oxygen concentration tracking error value, unit is mg / L; the blue solid line is the actual dissolved oxygen concentration value, the black solid line is the dissolved oxygen concentration setpoint, and the blue dashed line is the dissolved oxygen concentration tracking error value.

Claims

1. A smart robust control method for dissolved oxygen concentration based on a sliding mode observer, characterized in that: A predictive model for dissolved oxygen concentration in the biochemical reaction process of urban wastewater treatment based on an adaptive predictive fuzzy neural network is established. A disturbance information estimation model for the biochemical reaction process of urban wastewater treatment based on a sliding mode observer is constructed. An intelligent robust controller based on a sliding mode observer is designed to achieve accurate control of dissolved oxygen concentration. The process includes the following steps: (1) Extracting the characteristics of dissolved oxygen concentration changes in the biochemical reaction process of urban sewage treatment The dissolved oxygen concentration in the biochemical reaction process of urban wastewater treatment fluctuates significantly over time and is easily affected by changes in influent flow rate, water quality fluctuations, and operating environment. The dissolved oxygen concentration control model for the fifth zone of the biochemical reaction tank is as follows: Among them, S O,5 (t) represents the dissolved oxygen concentration in the fifth zone of the biochemical reaction tank at time t, h(S) O,5 (t) represents the change in dissolved oxygen concentration in the fifth zone of the biochemical reactor at time t, K L a5(t) represents the oxygen transfer coefficient of the fifth zone of the biochemical reactor at time t, d e (t) represents the external disturbances caused by changes in influent flow, water quality fluctuations, and operating environment during the biochemical reaction process of urban wastewater treatment at time t. d e The dynamic expression for (t) is: Where Q4(t) is the flow velocity in the fourth zone, and Q4(t) has different values ​​under different weather conditions, S O,4 This represents the dissolved oxygen concentration in the fourth zone of the biochemical reaction tank, with a value of 2 mg / L. (2) Establish a prediction model for dissolved oxygen concentration in the biochemical reaction process of urban sewage treatment. An adaptive fuzzy neural network prediction model was designed to predict the change in dissolved oxygen concentration h(S) in the fifth zone of the biochemical reaction tank during urban wastewater treatment. O,5 The prediction of (t) is achieved by an adaptive prediction fuzzy neural network with four layers: input layer, membership function layer, rule layer, and output layer. The network structure is 2-ll-1, where l is a positive integer greater than 1 and its initial value is 3. Specifically: Input layer: The input layer consists of 2 neurons, g(t) = [g1(t), g2(t)] T Let g1(t) be the input vector of the adaptive prediction fuzzy neural network at time t. O,5 g2(t) = K L a5(t), where T represents the transpose of the vector; Membership function layer: The membership function layer has l neurons, using a Gaussian function as the membership function. The output of the membership function layer is: Where, μ ij (t) represents the output value of the j-th neuron in the membership function layer at time t, i = 1, 2, j = 1, 2, ..., l, g i (t) represents the i-th variable in the input vector g(t) of the input layer, c ij (t) represents the center value of the i-th neuron in the input layer and the j-th neuron in the membership function layer at time t, a ij (t) represents the width values ​​of the i-th neuron in the input layer and the j-th neuron in the membership function layer at time t; Rule layer: The rule layer has l neurons, and the output of each neuron in the rule layer is: Where, γ j (t) represents the output value of the j-th neuron in the regular layer at time t; Output layer: The output of the output layer is the predicted value of the change in dissolved oxygen concentration in the fifth partition at time t. in, Let w(t) be the predicted value of the change in dissolved oxygen concentration in the fifth zone at time t, where w(t) = [w1(t), ..., w l [w(t)] represents the output weight vector of the adaptive prediction fuzzy neural network at time t, and w1(t) represents the output weight of the first neuron in the regular layer of the adaptive prediction fuzzy neural network at time t. l γ(t) represents the output weight of the l-th neuron in the rule layer of the adaptive prediction fuzzy neural network at time t, where γ(t) = [γ1(t), ..., γ2(t)]. l (t)] T Let γt be the output vector of the regular layer neuron in the adaptive prediction fuzzy neural network at time t, and let γ1(t) be the output value of the first neuron in the regular layer of the adaptive prediction fuzzy neural network at time t. l (t) represents the output value of the l-th neuron in the regular layer of the adaptive prediction fuzzy neural network at time t; Design a dissolved oxygen concentration predictor for the fifth zone of the urban wastewater treatment biochemical reaction process: in, S is the derivative of the estimated dissolved oxygen concentration in the fifth partition at time t. O,5d (t) represents the set value of dissolved oxygen concentration at time t, and ε(t) is the prediction error of dissolved oxygen concentration in the fifth zone at time t: in, This is the estimated value of the dissolved oxygen concentration in the fifth partition at time t; Using the dissolved oxygen concentration prediction error ε(t) in the fifth region, an adaptive law for the parameters of an adaptive prediction fuzzy neural network is designed: X j (t+1)=Ξ j (t)-0.1[γ j (t)|ε(t)|I+2Ξ j (t)] (8) Among them, Ξ j (t)=[w j (t),c j (t),a j [(t)] represents the parameter vector of the j-th neuron in the regular layer of the adaptive prediction fuzzy neural network at time t, w j (t) represents the output weights of the j-th neuron in the rule layer of the adaptive prediction fuzzy neural network at time t, and c j (t) represents the center of the j-th neuron in the regular layer of the adaptive prediction fuzzy neural network at time t, a j (t) represents the width of the j-th neuron in the regular layer of the adaptive prediction fuzzy neural network at time t, I represents a three-dimensional row vector of all 1s, Ξ j (t+1)=[w j (t+1),c j (t+1),a j [(t+1)] represents the parameter vector of the j-th neuron in the regular layer of the adaptive prediction fuzzy neural network at time t+1, w j (t+1) represents the output weight of the j-th neuron in the rule layer of the adaptive prediction fuzzy neural network at time t+1, c ij (t+1) represents the center value of the i-th neuron in the input layer and the j-th neuron in the membership function layer at time t+1, a ij (t+1) represents the width values ​​of the i-th neuron in the input layer and the j-th neuron in the membership function layer at time t+1; (3) Design a disturbance information estimation model for the biochemical reaction process of urban wastewater treatment. Construct a sliding mode observer-based model for estimating disturbance information in the biochemical reaction process of urban wastewater treatment to estimate external disturbances d caused by influent flow, water quality fluctuations, and changes in the operating environment. e The estimate of (t) is as follows: in, in, This represents the external perturbation d during the biochemical reaction process obtained through a sliding mode observer. e (t) is the derivative of the observed value. sgn() represents the sign function, and χ1(t) represents the S value during the biochemical reaction process obtained through the sliding mode observer. O,5 The observed value of χ²(t), where χ²(t) represents the first derivative of the external perturbation during the biochemical reaction obtained through the sliding mode observer. The observed values, This indicates that S represents the biochemical reaction process obtained through a sliding mode observer. O,5 (t) is the derivative of the observed value. This represents the external perturbation d during the biochemical reaction process obtained through a sliding mode observer. e (t) The second derivative of the observed value, according to formula (11) for external disturbances d in the biochemical reaction process. e (t) is estimated; (4) Design a smart robust control law for dissolved oxygen based on a sliding mode observer. ① At the initial time t=1 of the control action, the initial values ​​of the adaptive prediction fuzzy neural network parameters are randomly generated; ② Calculate the predicted value of the change in dissolved oxygen concentration in the fifth zone according to formula (5). ③ Solve the adaptive law of the adaptive prediction fuzzy neural network parameters according to formula (8); ④ Solve for the observed values ​​of influent flow rate, water quality fluctuations, and external disturbances caused by changes in the operating environment during the biochemical reaction process using formula (9). d (t); ⑤ Calculate the intelligent robust control law u(t) for dissolved oxygen in the sliding mode observer at time t: Calculate the dissolved oxygen concentration tracking and control error in the fifth zone: e(t)=S O,5d (t)-S O,5 (t) (12) Where e(t) represents the dissolved oxygen concentration setpoint S at time t. O,5d (t) and actual value S O,5 The error of (t); in, S represents the dissolved oxygen concentration setpoint at time t. O,5d (t) and actual value S O,5 (t) is the derivative of the error, and u(t) is the oxygen transfer coefficient of the controller at time t; The intelligent robust control law u(t) for dissolved oxygen at time t is calculated using formulas (5) and (11): in, This represents the derivative of the dissolved oxygen concentration setpoint at time t. ⑥ Increment time t by 1. If t < 200, return to step ②. If t = 200, end the loop. (5) Intelligent and robust control of dissolved oxygen concentration The dissolved oxygen concentration in the fifth zone of the urban wastewater treatment biochemical reaction process is tracked and controlled using the solved u(t). The output of the control system is the actual dissolved oxygen concentration value.

Images