Park water environment intelligent management system

Abstract

The invention discloses a park water environment intelligent management system, and relates to the field of water environment management. Comprising a plurality of groups of sensor modules which are uniformly deployed in a park water area, and the sensor module of each monitoring node comprises a pH sensor, an optical dissolved oxygen sensor, a COD sensor and an NH3-N sensor; the edge computing node is used for performing data fusion processing on the data acquired by the sensor module and caching the original data; the cloud analysis platform analyzes and processes the fused data, and a built-in data storage server is used for storing the data; and the execution terminal comprises a variable-frequency submersible sewage pump, a chemical dosing machine and an aeration machine. By means of front-end sensing, node processing, cloud analysis and rear-end execution, closed-loop monitoring, intelligent analysis and automatic processing of whole water quality monitoring and water quality processing are achieved, and therefore the management efficiency of the whole park water environment is improved.

Description

Technical Field

[0001] The present invention relates to the field of water environment management, and in particular to an intelligent management system for park water environment. Background Art

[0002] With the implementation of the concepts of industrial clustering, modularization and standardized management of industrial enterprises, various places will integrate similar industrial enterprises or upstream and downstream industrial enterprises to form clustered functional industrial parks. On the one hand, this facilitates enterprise management, and on the other hand, it also increases the interaction efficiency for relatively independent but business-related enterprises.

[0003] With the continuous acceleration of industrialization, the discharge of industrial wastewater continues to increase, causing serious environmental pollution. Wastewater contains a large number of harmful substances, such as heavy metals and organic matter, which pose a significant threat to aquatic ecosystems and human health. Therefore, the treatment of industrial wastewater is one of the current priorities of environmental protection.

[0004] Industrial wastewater treatment is of great significance. First, it effectively reduces water pollution, improves water quality, and protects public health and ecological safety. Second, it promotes sustainable industrial development, enhances a company's environmental image, and strengthens its market competitiveness. Furthermore, industrial wastewater treatment conserves water resources, improves water utilization efficiency, and reduces environmental protection costs.

[0005] Traditional park water environment management mainly relies on manual regular sampling and laboratory analysis, which has shortcomings such as monitoring lag, poor water treatment accuracy and low efficiency.

[0006] To this end, the present invention proposes an intelligent management system for park water environment. Summary of the Invention

[0007] The purpose of the present invention is to solve the shortcomings of the existing technology and to propose an intelligent management system for park water environment.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions:

[0009] An intelligent management system for park water environment, comprising:

[0010] Multiple sets of sensor modules are evenly deployed in the park's waters. The sensor modules of each monitoring node include pH sensors, optical dissolved oxygen sensors, COD sensors, and NH3-N sensors.

[0011] Edge computing nodes, which perform data fusion processing on the data collected by the sensor modules and cache the original data;

[0012] A cloud-based analysis platform that analyzes and processes the integrated data, and has a built-in data storage server for data storage;

[0013] The execution terminal includes a variable frequency submersible sewage pump, a chemical dosing machine and an aerator.

[0014] Preferably, the operation logic of the park water environment intelligent management system is:

[0015] S1: Multiple sensor modules collect node data in the park's waters and transmit the node data to edge computing nodes;

[0016] S2: The edge computing node processes the collected data and sends the processed data to the cloud analysis platform;

[0017] S3: A water quality evaluation model is built in the cloud analysis platform, and the received processed data is input into the model to generate water quality evaluation results;

[0018] S4: Dynamic optimization scheduling: The cloud-based analysis platform establishes a multi-objective optimization model based on the water quality evaluation results and solves the optimal result through constraints;

[0019] S5: Execute terminal control, generate control instructions according to the optimal result, and control the variable frequency submersible sewage pump, chemical dosing machine and aerator respectively according to the control instructions.

[0020] Preferably: in the step S2, the data processing is noise filtering, for the raw data collected by the i-th sensor at time t The filtering model is: in:

[0021] is the average value of the data in the past a minutes;

[0022] is the standard deviation of the data in the past a minutes;

[0023] a is the length of the data comparison period, a=30 or 60 or 90;

[0024] when Greater than |δ z |When the data is determined to be an outlier, data repair is performed.

[0025] Preferably: in the step S2, z is the discrete standard threshold, Where b is the length of the reference time period.

[0026] Preferably: in step S2, the formula for data repair is: z′ t Replace the value with the repaired data.

[0027] Preferably: in the step S3, the expression of the water quality evaluation model is: Among them, P i is the score of the i-th dimension of water pollution, w i is the i-th scoring dimension, and n is the total number of dimensions.

[0028] Preferably: in the step S3, the score P of the i-th dimension i The calculation formula is: in( i M0, i , M′0) is the corresponding interval of the detection value of the i-th dimension when the water quality is optimal.

[0029] Preferably: in step S4, the objective function of the multi-objective optimization model is: in:

[0030] Y is the cost of raw materials required to change the unit percentage of water quality;

[0031] Q is the energy cost when changing the water quality unit percentage, Q=(e1q 3 t +e2q 2 t +e3q t ), e1, e2, e3 are the characteristic curve coefficients of the pump, q t It is the real-time flow output of the pump;

[0032] C is the penalty cost when the water quality score P deviates from the optimal water quality score P0 by a unit percentage;

[0033] α and β are weights.

[0034] Preferably: in the step S4, the constraint condition is is the rated flow rate of the pump.

[0035] Preferably, in step S5, the control adopts a compensatory control form, which specifically includes the following steps:

[0036] S51: Obtain the previous theoretical control command value L and the actual execution value L′, and calculate the difference ΔL=LL′;

[0037] S52: Establish an execution error model based on time length, Among them, ΔL0 is the error prediction value of this execution, L u and L u ′ is the theoretical instruction control value and actual execution value of the uth execution forward from the current time, w′ u is the importance of participation in the uth time, w′ u >w′ u+1 ;

[0038] S53: When the theoretical control value is L0, it is updated to L0-ΔL0 according to the error compensation.

[0039] The beneficial effects of the present invention are:

[0040] 1. The present invention realizes closed-loop monitoring, intelligent analysis and automated processing of the entire water quality monitoring and treatment through front-end perception, node processing, cloud analysis and back-end execution, thereby increasing the management efficiency of the water environment of the entire park.

[0041] 2. The present invention first uses discreteness to judge the collected data of the sensor. Based on the abnormality judgment, a piecewise function is used to compensate the abnormal data based on the mean and standard deviation according to the situation of discreteness offset, thereby increasing the accuracy of data collection and laying the foundation for subsequent data analysis.

[0042] 3. The present invention uses a scoring form to evaluate water quality, and uses a piecewise function to evaluate the water quality deviation percentage using the optimal interval of the dimensional data corresponding to the water quality, thereby increasing the accuracy of the water quality evaluation and providing a solid foundation for the subsequent execution of water treatment control.

[0043] 4. The present invention adopts a cost-based objective function minimum solution method for water quality optimization, and the cost takes into account the cost of chemicals, energy consumption costs and penalty costs for exceeding emission standards, thereby significantly reducing the water treatment cost of enterprises while ensuring comprehensive consideration.

[0044] 5. The present invention adopts compensatory control for the control of the execution end, thereby increasing the control accuracy, and adopts the form of weighted summation of comparison of past errors for the compensatory control, and the weight of the error value closer to the current control time is greater, so that the error compensation is more consistent with the intake working condition of the execution end, making the compensation more reasonable and accurate, thereby further increasing the control accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is an architecture diagram of a park water environment intelligent management system proposed by the present invention;

[0046] Figure 2 This is a processing logic diagram of the park water environment intelligent management system proposed by the present invention. DETAILED DESCRIPTION

[0047] The technical solution of the present invention will be further described in detail below in conjunction with specific implementation methods.

[0048] Example 1:

[0049] A park water environment intelligent management system, comprising:

[0050] Multiple sets of sensor modules are evenly deployed in the park's waters. The sensor modules of each monitoring node include pH sensors, optical dissolved oxygen sensors, COD sensors, and NH3-N sensors.

[0051] Edge computing nodes, which perform data fusion processing on the data collected by the sensor modules and cache the original data;

[0052] A cloud-based analysis platform that analyzes and processes the integrated data, and has a built-in data storage server for data storage;

[0053] The execution terminal includes a variable frequency submersible sewage pump, a chemical dosing machine and an aerator.

[0054] The operating logic of the park water environment intelligent management system is:

[0055] S1: Multiple sensor modules collect node data in the park's waters and transmit the node data to edge computing nodes;

[0056] S2: The edge computing node processes the collected data and sends the processed data to the cloud analysis platform;

[0057] S3: A water quality evaluation model is built in the cloud analysis platform, and the received processed data is input into the model to generate water quality evaluation results;

[0058] S4: Dynamic optimization scheduling: The cloud-based analysis platform establishes a multi-objective optimization model based on the water quality evaluation results and solves the optimal result through constraints;

[0059] S5: Execute terminal control, generate control instructions according to the optimal result, and control the variable frequency submersible sewage pump, chemical dosing machine and aerator respectively according to the control instructions.

[0060] The present invention realizes closed-loop monitoring, intelligent analysis and automated processing of the entire water quality monitoring and treatment through front-end perception, node processing, cloud analysis and back-end execution, thereby increasing the management efficiency of the water environment of the entire park.

[0061] Example 2:

[0062] A park water environment intelligent management system, comprising:

[0063] Multiple sets of sensor modules are evenly deployed in the park's waters. The sensor modules of each monitoring node include pH sensors, optical dissolved oxygen sensors, COD sensors, and NH3-N sensors.

[0064] Edge computing nodes, which perform data fusion processing on the data collected by the sensor modules and cache the original data;

[0065] A cloud-based analysis platform that analyzes and processes the integrated data, and has a built-in data storage server for data storage;

[0066] The execution terminal includes a variable frequency submersible sewage pump, a chemical dosing machine and an aerator.

[0067] The operating logic of the park water environment intelligent management system is:

[0068] S1: Multiple sensor modules collect node data in the park's waters and transmit the node data to edge computing nodes;

[0069] S2: The edge computing node processes the collected data and sends the processed data to the cloud analysis platform;

[0070] S3: A water quality evaluation model is built in the cloud analysis platform, and the received processed data is input into the model to generate water quality evaluation results;

[0071] S4: Dynamic optimization scheduling: The cloud-based analysis platform establishes a multi-objective optimization model based on the water quality evaluation results and solves the optimal result through constraints;

[0072] S5: Execute terminal control, generate control instructions according to the optimal result, and control the variable frequency submersible sewage pump, chemical dosing machine and aerator respectively according to the control instructions.

[0073] In the step S2, the data processing is noise filtering. For the raw data collected by the i-th sensor at time t, The filtering model is: in:

[0074] is the average value of the data in the past a minutes;

[0075] is the standard deviation of the data in the past a minutes;

[0076] a is the length of the data comparison period, a=30 or 60 or 90;

[0077] when Greater than |δ z |When the data is determined to be an outlier, data repair is performed.

[0078] In the S2 step, δ z is the discrete standard threshold, Where b is the length of the reference time period.

[0079] In the step S2, the formula for data repair is: z′ t Replace the value with the repaired data.

[0080] The present invention first uses the discreteness to judge the collected data of the sensor. On the basis of judging the abnormality, a piecewise function is used to compensate the abnormal data based on the mean and standard deviation according to the situation of discreteness offset, thereby increasing the accuracy of data collection and laying the foundation for subsequent data analysis.

[0081] Example 3:

[0082] A park water environment intelligent management system, comprising:

[0083] Multiple sets of sensor modules are evenly deployed in the park's waters. The sensor modules of each monitoring node include pH sensors, optical dissolved oxygen sensors, COD sensors, and NH3-N sensors.

[0084] Edge computing nodes, which perform data fusion processing on the data collected by the sensor modules and cache the original data;

[0085] A cloud-based analysis platform that analyzes and processes the integrated data, and has a built-in data storage server for data storage;

[0086] The execution terminal includes a variable frequency submersible sewage pump, a chemical dosing machine and an aerator.

[0087] The operating logic of the park water environment intelligent management system is:

[0088] S1: Multiple sensor modules collect node data in the park's waters and transmit the node data to edge computing nodes;

[0089] S2: The edge computing node processes the collected data and sends the processed data to the cloud analysis platform;

[0090] S3: A water quality evaluation model is built in the cloud analysis platform, and the received processed data is input into the model to generate water quality evaluation results;

[0091] S4: Dynamic optimization scheduling: The cloud-based analysis platform establishes a multi-objective optimization model based on the water quality evaluation results and solves the optimal result through constraints;

[0092] S5: Execute terminal control, generate control instructions according to the optimal result, and control the variable frequency submersible sewage pump, chemical dosing machine and aerator respectively according to the control instructions.

[0093] In the step S2, the data processing is noise filtering. For the raw data collected by the i-th sensor at time t, The filtering model is: in:

[0094] is the average value of the data in the past a minutes;

[0095] is the standard deviation of the data in the past a minutes;

[0096] a is the length of the data comparison period, a=30 or 60 or 90;

[0097] when Greater than |δ z |When the data is determined to be an outlier, data repair is performed.

[0098] In the S2 step, δ z is the discrete standard threshold, Where b is the length of the reference time period.

[0099] In the step S2, the formula for data repair is: z′ t Replace the value with the repaired data.

[0100] In step S3, the expression of the water quality evaluation model is: Among them, P i is the score of the i-th dimension of water pollution, w i is the i-th scoring dimension, n is the total number of dimensions, and in this embodiment, since four sensors are used, n=4, P1-P n Score the pollution levels collected by the four sensors respectively.

[0101] In the S3 step, the score P of the i-th dimension i The calculation formula is: in( i M0, i M0′) is the corresponding interval of the detection value of the i-th dimension when the water quality is optimal.

[0102] The present invention adopts the form of scoring for water quality evaluation, and uses a piecewise function to evaluate the water quality deviation percentage using the optimal interval of the dimensional data corresponding to the water quality, thereby increasing the accuracy of the water quality evaluation and providing a solid foundation for the subsequent execution of water treatment control.

[0103] Example 4:

[0104] A park water environment intelligent management system, comprising:

[0105] Multiple sets of sensor modules are evenly deployed in the park's waters. The sensor modules of each monitoring node include pH sensors, optical dissolved oxygen sensors, COD sensors, and NH3-N sensors.

[0106] Edge computing nodes, which perform data fusion processing on the data collected by the sensor modules and cache the original data;

[0107] A cloud-based analysis platform that analyzes and processes the integrated data, and has a built-in data storage server for data storage;

[0108] The execution terminal includes a variable frequency submersible sewage pump, a chemical dosing machine and an aerator.

[0109] The operating logic of the park water environment intelligent management system is:

[0110] S1: Multiple sensor modules collect node data in the park's waters and transmit the node data to edge computing nodes;

[0111] S2: The edge computing node processes the collected data and sends the processed data to the cloud analysis platform;

[0112] S3: A water quality evaluation model is built in the cloud analysis platform, and the received processed data is input into the model to generate water quality evaluation results;

[0113] S4: Dynamic optimization scheduling: The cloud-based analysis platform establishes a multi-objective optimization model based on the water quality evaluation results and solves the optimal result through constraints;

[0114] S5: Execute terminal control, generate control instructions according to the optimal result, and control the variable frequency submersible sewage pump, chemical dosing machine and aerator respectively according to the control instructions.

[0115] In the step S2, the data processing is noise filtering. For the raw data collected by the i-th sensor at time t, The filtering model is: in:

[0116] is the average value of the data in the past a minutes;

[0117] is the standard deviation of the data in the past a minutes;

[0118] a is the length of the data comparison period, a=30 or 60 or 90;

[0119] when Greater than |δ z |When the data is determined to be an outlier, data repair is performed.

[0120] In the S2 step, δ z is the discrete standard threshold, Where b is the length of the reference time period.

[0121] In the step S2, the formula for data repair is: z′ t Replace the value with the repaired data.

[0122] In step S3, the expression of the water quality evaluation model is: Among them, P i is the score of the i-th dimension of water pollution, w i is the i-th scoring dimension, n is the total number of dimensions. In this embodiment, since four sensors are used, n=4, P1-P n Score the pollution levels collected by the four sensors respectively.

[0123] In the S3 step, the score P of the i-th dimension i The calculation formula is: in( i M0, i M0′) is the corresponding interval of the detection value of the i-th dimension when the water quality is optimal.

[0124] In step S4, the objective function of the multi-objective optimization model is: in:

[0125] Y is the cost of raw materials required to change the water quality by a unit percentage (e.g., if adding a chemical, Y is the purchase cost of the chemical when the water quality changes by 1%. If aeration is used, the cost is 0 when air is used, and the cost of other gases when other gases are used).

[0126] Q is the energy cost when changing the water quality unit percentage, Q=(e1q 3 t +e2q 2 t +e3q t ), e1, e2, e3 are the characteristic curve coefficients of the pump, which are calibrated by the manufacturer's experiment, q t is the real-time flow output of the pump (e.g. when adding chemicals, Q is the energy consumption of the water pump when the water quality score changes by 1%, and if it is a change in aeration, it is the energy consumption of the air pump);

[0127] C is the penalty cost when the water quality score P deviates from the optimal water quality score P0 by a unit percentage, which is determined according to the penalty policy of the Environmental Regulatory Bureau;

[0128] α and β are weights.

[0129] In the step S4, the constraints are is the rated flow rate of the pump.

[0130] The present invention adopts a cost-based objective function minimum solution method for water quality optimization, and takes into account the cost of chemicals, energy consumption costs and penalty costs for exceeding emission standards, thereby significantly reducing the water treatment cost of enterprises while ensuring comprehensive consideration.

[0131] Example 5:

[0132] A park water environment intelligent management system, comprising:

[0133] Multiple sets of sensor modules are evenly deployed in the park's waters. The sensor modules of each monitoring node include pH sensors, optical dissolved oxygen sensors, COD sensors, and NH3-N sensors.

[0134] Edge computing nodes, which perform data fusion processing on the data collected by the sensor modules and cache the original data;

[0135] A cloud-based analysis platform that analyzes and processes the integrated data, and has a built-in data storage server for data storage;

[0136] The execution terminal includes a variable frequency submersible sewage pump, a chemical dosing machine and an aerator.

[0137] The operating logic of the park water environment intelligent management system is:

[0138] S1: Multiple sensor modules collect node data in the park's waters and transmit the node data to edge computing nodes;

[0139] S2: The edge computing node processes the collected data and sends the processed data to the cloud analysis platform;

[0140] S3: A water quality evaluation model is built in the cloud analysis platform, and the received processed data is input into the model to generate water quality evaluation results;

[0141] S4: Dynamic optimization scheduling: The cloud-based analysis platform establishes a multi-objective optimization model based on the water quality evaluation results and solves the optimal result through constraints;

[0142] S5: Execute terminal control, generate control instructions according to the optimal result, and control the variable frequency submersible sewage pump, chemical dosing machine and aerator respectively according to the control instructions.

[0143] In the step S2, the data processing is noise filtering. For the raw data collected by the i-th sensor at time t, The filtering model is: in:

[0144] is the average value of the data in the past a minutes;

[0145] is the standard deviation of the data in the past a minutes;

[0146] a is the length of the data comparison period, a=30 or 60 or 90;

[0147] when Greater than |δ z |When the data is determined to be an outlier, data repair is performed.

[0148] In the S2 step, δ z is the discrete standard threshold, Where b is the length of the reference time period.

[0149] In the step S2, the formula for data repair is: z′ t Replace the value with the repaired data.

[0150] In step S3, the expression of the water quality evaluation model is: Among them, P i is the score of the i-th dimension of water pollution, w i is the i-th scoring dimension, n is the total number of dimensions, and in this embodiment, since four sensors are used, n=4, P1-P n Score the pollution levels collected by the four sensors respectively.

[0151] In the S3 step, the score P of the i-th dimension i The calculation formula is: in( i M0, i M0′) is the corresponding interval of the detection value of the i-th dimension when the water quality is optimal.

[0152] In step S4, the objective function of the multi-objective optimization model is: in:

[0153] Y is the cost of raw materials required to change the water quality by a unit percentage (e.g., if adding a chemical, Y is the purchase cost of the chemical when the water quality changes by 1%. If aeration is used, the cost is 0 when air is used, and the cost of other gases when other gases are used).

[0154] Q is the energy cost when changing the water quality unit percentage, Q=(e1q 3 t +e2q 2 t +e3q t), e1, e2, e3 are the characteristic curve coefficients of the pump, which are calibrated by the manufacturer's experiment, q t is the real-time flow output of the pump (e.g. when adding chemicals, Q is the energy consumption of the water pump when the water quality score changes by 1%, and if it is a change in aeration, it is the energy consumption of the air pump);

[0155] C is the penalty cost when the water quality score P deviates from the optimal water quality score P0 by a unit percentage, which is determined according to the penalty policy of the Environmental Regulatory Bureau;

[0156] α and β are weights.

[0157] In the step S4, the constraints are is the rated flow rate of the pump.

[0158] In step S5, the control adopts a compensatory control form, which specifically includes the following steps:

[0159] S51: Obtain the previous theoretical control command value L and the actual execution value L′, and calculate the difference ΔL=LL′;

[0160] S52: Establish an execution error model based on time length, Among them, ΔL0 is the error prediction value of this execution, L u and L u ′ is the theoretical instruction control value and actual execution value of the uth execution forward from the current time, w′ u is the importance of participation in the uth time, w′ u >w′ u+1 ;

[0161] S53: When the theoretical control value is L0, it is updated to L0-ΔL0 according to the error compensation.

[0162] The present invention adopts compensatory control for the control of the execution end, thereby increasing the control accuracy, and adopts the form of weighted summation of comparison of past errors for the compensatory control, and the weight of the error value closer to the current control time is greater, so that the error compensation is more consistent with the intake working condition of the execution end, making the compensation more reasonable and accurate, thereby further increasing the control accuracy.

[0163] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A park water environment intelligent management system, characterized in that: include: Multiple sets of sensor modules are evenly deployed in the park's waters. The sensor modules of each monitoring node include pH sensors, optical dissolved oxygen sensors, COD sensors, and NH3-N sensors. Edge computing nodes, which perform data fusion processing on the data collected by the sensor modules and cache the original data; A cloud-based analysis platform that analyzes and processes the integrated data, and has a built-in data storage server for data storage; The execution terminal includes a variable frequency submersible sewage pump, a chemical dosing machine and an aerator.

2. A park water environment intelligent management system according to claim 1, characterized in that: The operating logic of the park water environment intelligent management system is: S1: Multiple sensor modules collect node data in the park's waters and transmit the node data to edge computing nodes; S2: The edge computing node processes the collected data and sends the processed data to the cloud analysis platform; S3: A water quality evaluation model is built in the cloud analysis platform, and the received processed data is input into the model to generate water quality evaluation results; S4: Dynamic optimization scheduling: The cloud-based analysis platform establishes a multi-objective optimization model based on the water quality evaluation results and solves the optimal result through constraints; S5: Execute terminal control, generate control instructions according to the optimal result, and control the variable frequency submersible sewage pump, chemical dosing machine and aerator respectively according to the control instructions.

3. A park water environment intelligent management system according to claim 2, characterized in that: In the step S2, the data processing is noise filtering. For the raw data collected by the i-th sensor at time t, The filtering model is: in: is the average value of the data in the past a minutes; is the standard deviation of the data in the past a minutes; a is the length of the data comparison period, a=30 or 60 or 90; when Greater than |δ z |When the data is determined to be an outlier, data repair is performed.

4. A park water environment intelligent management system according to claim 3, characterized in that: In the S2 step, δ z is the discrete standard threshold, Where b is the length of the reference time period.

5. The park water environment intelligent management system according to claim 4, characterized in that: In the step S2, the formula for data repair is: z′ t Replace the value with the repaired data.

6. The park water environment intelligent management system according to claim 2, characterized in that: In step S3, the expression of the water quality evaluation model is: Among them, P i is the score of the i-th dimension of water pollution, w i is the i-th scoring dimension, and n is the total number of dimensions.

7. The park water environment intelligent management system according to claim 6, characterized in that: In the S3 step, the score P of the i-th dimension i The calculation formula is: in( i M0, i M0′) is the corresponding interval of the detection value of the i-th dimension when the water quality is optimal.

8. The park water environment intelligent management system according to claim 6, characterized in that: In step S4, the objective function of the multi-objective optimization model is: in: Y is the cost of raw materials required to change the unit percentage of water quality; Q is the energy cost when changing the water quality unit percentage, Q=(e1q 3 t +e2q 2 t +e3q t ), e1, e2, e3 are the characteristic curve coefficients of the pump, q t It is the real-time flow output of the pump; C is the penalty cost when the water quality score P deviates from the optimal water quality score P0 by a unit percentage; α and β are weights.

9. The park water environment intelligent management system according to claim 8, characterized in that: In the step S4, the constraints are is the rated flow rate of the pump.

10. The park water environment intelligent management system according to claim 2, characterized in that: In step S5, the control adopts a compensatory control form, which specifically includes the following steps: S51: Obtain the previous theoretical control command value L and the actual execution value L′, and calculate the difference ΔL=LL′; S52: Establish an execution error model based on time length, Among them, ΔL0 is the error prediction value of this execution, L u and L u ′ is the theoretical instruction control value and actual execution value of the uth execution forward from the current time, w′ u is the importance of participation in the uth time, w′ u >w′ u+1 ; S53: When the theoretical control value is L0, it is updated to L0-ΔL0 according to the error compensation.

Images