A method and system for collaborative optimization of energy consumption and operation and maintenance cost of a filtration device

Abstract

The present application relates to the technical field of aquaculture management, and particularly relates to a filter device energy consumption and operation and maintenance cost collaborative optimization method and system, which integrates three types of core data of equipment operation, water quality and cost, builds independent functions of energy consumption and operation and maintenance cost, introduces weight coefficients, nonlinear interaction terms and dynamic marginal effect terms based on water quality data, establishes a coupled optimization model, and breaks the single optimization limitation. Through an improved NSGA-III algorithm, bidirectional pressure reduction of energy consumption and operation and maintenance cost is realized, equipment failure caused by low energy consumption and rapid increase of operation and maintenance cost are avoided, energy waste caused by excessive pursuit of operation and maintenance convenience is eliminated, and the economic efficiency and sustainability of the aquaculture system are significantly improved, thereby providing key technical support for efficient, low-carbon and intelligent transformation of the aquaculture industry.

Description

Technical Field

[0001] This invention relates to the field of aquaculture management technology, and in particular to a method and system for synergistic optimization of energy consumption and operation and maintenance costs of filtration devices. Background Technology

[0002] Against the backdrop of large-scale and intensive development in aquaculture, filtration devices, as core equipment for maintaining the cleanliness of aquaculture water and ensuring the health of aquaculture organisms, have seen their energy consumption and operation and maintenance costs become key factors restricting the improvement of aquaculture efficiency.

[0003] Existing filtration devices generally suffer from a disconnect between energy consumption and operation and maintenance management. Most optimizations are made solely for energy consumption (such as variable frequency pump adjustment) or operation and maintenance costs (such as filter material replacement cycle), ignoring the coupling contradiction between the two: simply pursuing low energy consumption can easily lead to decreased filtration efficiency, frequent equipment failures, and increased operation and maintenance labor and parts replacement costs; while overemphasizing operation and maintenance convenience may result in energy waste and drive up long-term operating costs.

[0004] Therefore, there is an urgent need for a synergistic optimization scheme for the energy consumption and operation and maintenance costs of filtration devices in aquaculture, so as to achieve a two-way reduction in energy consumption and operation and maintenance costs, and improve the economy and sustainability of the aquaculture system. This is of great practical significance for promoting the transformation of the aquaculture industry towards high efficiency, low carbon and intelligentization. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method and system for synergistic optimization of energy consumption and operation and maintenance costs of a filtration device, which can achieve a bidirectional reduction in energy consumption and operation and maintenance costs.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0007] A method for synergistic optimization of energy consumption and operation and maintenance costs of a filtration device includes the following steps:

[0008] S1. Collect equipment operation data of the filtration equipment, collect water quality data synchronously using online sensors installed in the aquaculture pond, and connect to the enterprise management system to obtain cost data;

[0009] S2. Construct a function representation of energy consumption based on the pump power, pump running time, backwash pump power, backwash time, disinfection equipment power, and disinfection equipment running time in the equipment operation data;

[0010] S3. Based on the energy consumption, electricity price, maintenance hours, hourly labor cost, filter media replacement frequency, filter media purchase cost, equipment purchase cost, depreciation rate, downtime due to failure, and the aquaculture loss per unit time due to downtime due to failure, construct a functional representation of the operation and maintenance cost;

[0011] S4. Assign weight coefficients to the energy consumption and the operation and maintenance cost, and introduce nonlinear interaction terms and dynamic marginal effect terms to construct a nonlinear coupled optimization model of the energy consumption and the operation and maintenance cost, and construct constraints.

[0012] The weighting coefficients for energy consumption and operation and maintenance costs are determined based on the water quality data.

[0013] S5. The nonlinear coupling optimization model is optimized and solved using the improved NSGA-Ⅲ algorithm, and the filtering equipment is optimized and controlled based on the solution results.

[0014] To solve the above-mentioned technical problems, another technical solution adopted by the present invention is as follows:

[0015] A system for collaboratively optimizing the energy consumption and operation and maintenance costs of a filtration device includes:

[0016] The data acquisition module collects equipment operation data from the filtration equipment, synchronously collects water quality data using online sensors installed in the aquaculture pond, and connects to the enterprise management system to obtain cost data.

[0017] The model building module constructs a functional representation of energy consumption based on the pump power, pump runtime, backwash pump power, backwash duration, disinfection equipment power, and disinfection equipment runtime in the equipment operation data.

[0018] Based on the energy consumption, electricity price per unit, maintenance hours, hourly labor cost, filter media replacement frequency, filter media purchase cost, equipment purchase cost, depreciation rate, downtime due to failure, and the aquaculture loss per unit time due to downtime due to failure, a functional representation of operation and maintenance cost is constructed.

[0019] Weighting coefficients are assigned to the energy consumption and the operation and maintenance cost, and nonlinear interaction terms and dynamic marginal effect terms are introduced to construct a nonlinear coupled optimization model of the energy consumption and the operation and maintenance cost, and constraints are constructed.

[0020] The weighting coefficients for energy consumption and operation and maintenance costs are determined based on the water quality data.

[0021] The model solving module optimizes and solves the nonlinear coupling optimization model using the improved NSGA-Ⅲ algorithm, and optimizes and controls the filtering device based on the solution results.

[0022] The beneficial effects of this invention are as follows: The present invention provides a method and system for the coordinated optimization of energy consumption and operation and maintenance costs of a filtration device. It integrates three core data categories: equipment operation, water quality, and cost. It constructs independent functions for energy consumption and operation and maintenance costs, and introduces weighting coefficients, nonlinear interaction terms, and dynamic marginal effect terms based on water quality data to establish a coupled optimization model, breaking the limitations of single optimization. Through the improved NSGA-Ⅲ algorithm, it achieves a two-way reduction in energy consumption and operation and maintenance costs. This avoids frequent equipment failures and soaring operation and maintenance costs caused by low energy consumption, and also eliminates energy waste caused by excessive pursuit of convenient operation and maintenance. It significantly improves the economy and sustainability of aquaculture systems, providing key technical support for the efficient, low-carbon, and intelligent transformation of the aquaculture industry. Attached Figure Description

[0023] Figure 1 This is a flowchart of a method for collaborative optimization of energy consumption and operation and maintenance costs of a filtration device according to an embodiment of the present invention;

[0024] Figure 2 This is a schematic diagram of a filtration device energy consumption and operation and maintenance cost collaborative optimization system according to an embodiment of the present invention;

[0025] Label Explanation:

[0026] 1. A system for collaborative optimization of energy consumption and operation and maintenance costs of a filtration device; 2. Data acquisition module; 3. Model building module; 4. Model solving module. Detailed Implementation

[0027] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0028] Please refer to Figure 1 One embodiment of the present invention is as follows:

[0029] A method for synergistic optimization of energy consumption and operation and maintenance costs of a filtration device includes the following steps:

[0030] S1. Collect equipment operation data of the filtration equipment, synchronously collect water quality data using online sensors installed in the aquaculture pond, and connect to the enterprise management system to obtain cost data.

[0031] In this embodiment, a water quality sensor group (suspended solids concentration, ammonia nitrogen, dissolved oxygen, pH sensor), an equipment operation monitoring module (current transformer, pressure transmitter, flow sensor), and a cost data interface unit are installed in the aquaculture pond, the filtration device, and the operation and maintenance terminal, respectively, to realize the real-time collection of water quality parameters (5 minutes / time), equipment operation data (5 minutes / time), and full-cycle cost data.

[0032] Specifically, online sensor arrays can be installed at the outlet of the aquaculture pond, the inlet of the filter device, and the outlet of the filter device. The specific parameters are as follows: suspended solids concentration sensor (measurement range 0-500 mg / L, accuracy ±5%), ammonia nitrogen sensor (measurement range 0-10 mg / L, accuracy ±0.1 mg / L), dissolved oxygen sensor (measurement range 0-20 mg / L, accuracy ±0.1 mg / L), and pH sensor (measurement range 4-10, accuracy ±0.1). The sensors are connected to an edge computing gateway via a 485 bus, with a fixed acquisition frequency of 5 minutes per acquisition to ensure the timeliness and continuity of water quality data.

[0033] Monitoring modules are installed on the main water pump, backwash pump, filter bed, and ultraviolet disinfection equipment of the filtration unit: a current transformer (accuracy ±1%) is installed at the main water pump to collect pump power in real time; pressure transmitters (measurement range 0-1MPa, accuracy ±0.01MPa) are installed at the inlet and outlet of the filter bed to monitor the pressure difference; flow sensors (measurement range 0-2m / s, accuracy ±0.05m / s) are installed in the filter pipeline to collect water flow velocity; and timing modules are installed on all equipment to record the cumulative running time. The data acquisition frequency is synchronized with the water quality acquisition nodes to achieve real-time monitoring of equipment status.

[0034] By integrating with the enterprise ERP system through the operation and maintenance management system, static cost data is automatically synchronized: electricity price (differentiated by peak, off-peak, and normal periods: peak period 1.0 yuan / kWh, normal period 0.6 yuan / kWh, off-peak period 0.3 yuan / kWh), labor maintenance hour cost (50 yuan / h), filter material purchase price (PP cotton filter material 80 yuan / ㎡, activated carbon filter material 12 yuan / ㎡), and equipment depreciation rate (based on the straight-line method, annual depreciation rate 10%). Simultaneously, dynamic cost data is entered in real time through the fault recording module: fault occurrence time, repair duration, replacement parts cost, and downtime aquaculture loss (calculated based on aquaculture density; downtime loss for Litopenaeus vannamei aquaculture is 200 yuan / h).

[0035] The collected data is processed using a three-step method: outlier removal, data standardization, and time series alignment, to ensure data quality.

[0036] Outlier removal: The 3σ criterion is used to identify and remove outlier data. For example, when the suspended solids concentration data exceeds the mean ± 3σ range, it is determined to be sensor fault data, and the data is supplemented by using valid data from adjacent time points through linear interpolation. For equipment operation data, when the pump power suddenly changes by more than 30%, it is determined to be an abnormal operating condition, marked, and included in subsequent fault analysis.

[0037] Data standardization: Min-max standardization is performed on data of different dimensions to map them to the [0,1] interval, thereby eliminating the impact of dimension differences on model calculation.

[0038] Time series alignment: Water quality data, equipment operation data, and cost data are aligned by timestamps with a 5-minute time granularity to form a dataset in a unified format. This dataset is stored in the local SQLite database of the edge computing gateway and simultaneously uploaded to the cloud server for backup. The data retention period is 1 year.

[0039] S2. Construct a function representation of energy consumption based on the pump power, pump running time, backwash pump power, backwash time, disinfection equipment power, and disinfection equipment running time in the equipment operation data;

[0040] energy consumption E The function is specifically represented as follows:

[0041] E = P pump × t pump + P backwash × t backwash + P disinfection × t disinfection ;

[0042] in, P pump Indicates the power of the water pump. t pump Indicates the operating time of the water pump. P backwash Indicates the power of the backwash water pump. t backwash Indicates the backwash duration. P disinfection Indicates the power of the disinfection equipment. t disinfection This indicates the operating time of the disinfection equipment.

[0043] S3. Based on the energy consumption, electricity price, maintenance hours, hourly labor cost, filter media replacement frequency, filter media purchase cost, equipment purchase cost, depreciation rate, downtime due to failure, and the aquaculture loss per unit time due to downtime due to failure, construct a functional representation of the operation and maintenance cost;

[0044] The operation and maintenance costs C The function is specifically represented as follows:

[0045] C = C electrity + C labor + C filter + Cdepreciation + C failure ;

[0046] C electrity = E × U ;

[0047] C labor = T × R ;

[0048] C filter = N × P filter ;

[0049] C depreciation = P depreciation × r ;

[0050] C failure = L × t failure ;

[0051] in, C electrity Indicates electricity cost. U This indicates the unit price of electricity. E Indicates energy consumption. C labor Indicates maintenance costs, T Indicates maintenance time. R This represents hourly labor costs. C filter Indicates the cost of filter media. N Indicates the number of times the filter media has been replaced. P filter This indicates the cost of a single filter material purchase. C depreciation This represents the cost of equipment depreciation. P depreciation This indicates the cost of purchasing the equipment. r Indicates the depreciation rate. C failure Indicates the cost of failure and damage. L Indicates the duration of downtime due to a fault. t failure This represents the aquaculture loss per unit of time caused by the downtime due to a malfunction.

[0052] S4. Assign weight coefficients to the energy consumption and the operation and maintenance cost, and introduce nonlinear interaction terms and dynamic marginal effect terms to construct a nonlinear coupled optimization model of the energy consumption and the operation and maintenance cost, and construct constraints.

[0053] The weighting coefficients for energy consumption and operation and maintenance costs are determined based on the water quality data.

[0054] energy consumption E With the aforementioned operation and maintenance costs C The nonlinear coupling optimization model is expressed as:

[0055] ;

[0056] in, F Represents the fitness value. k 1. k 2 represents the energy consumption respectively. E With the aforementioned operation and maintenance costs C The basic target weights are obtained based on fuzzy inference from the water quality data. k 1+ k 2=1, k 3 indicates the coefficient of the nonlinear interaction term, which is dynamically adjusted according to the equipment's operating status. k 4 represents the dynamic marginal effect term.

[0057] The relationship between energy consumption and operation and maintenance costs is not a linear additive one, but rather involves a threshold effect (such as an exponential increase in operation and maintenance costs after energy consumption decreases to a certain critical point) and a lag effect (such as a surge in long-term operation and maintenance costs due to short-term energy-saving operations). Linear models cannot characterize this non-linear feature. Furthermore, linear weighting only achieves a "formal balance" and fails to realize the synergistic effect of "energy consumption reduction driving cost optimization" or "cost control forcing energy efficiency improvement" in a mechanistic way.

[0058] In this embodiment, nonlinear interaction terms and dynamic marginal effect terms are introduced. Through a three-layer architecture of "nonlinear interaction terms + dynamic marginal effect coefficients + scenario constraint coupling", a coupled model that can accurately characterize the intrinsic relationship between energy consumption (E) and operation and maintenance costs (C) is constructed, realizing the upgrade from "formal coupling" to "mechanism coupling".

[0059] The nonlinear interaction term characterizes the inherent inverse trade-off between energy consumption and operation and maintenance costs. For example, reducing pump speed reduces energy consumption but accelerates filter media clogging, increases backwashing frequency, and consequently raises operation and maintenance costs; conversely, increasing pump speed delays filter media clogging but increases energy consumption. This nonlinear relationship, where a decrease in one leads to an increase in the other, is quantified through the product term of E and C. When both E and C are at low levels, the value of E×C is small, resulting in a weak penalty for F; when one increases significantly, the value of E×C increases rapidly, strengthening the penalty for F and forcing the algorithm to find a balance between the two.

[0060] The dynamic marginal effect term ensures the synergistic effect of "synchronous optimization." Optimization of energy consumption and operation and maintenance costs should be dynamically synchronized to avoid short-term behaviors such as "rapidly decreasing energy consumption but sharply increasing operation and maintenance costs." For example, reducing the number of backwashing cycles may lower short-term operation and maintenance costs, but this can lead to increased filter media clogging and a significant long-term increase in energy consumption. By calculating the difference in the time-varying rates of energy consumption and cost, penalties are imposed on "asynchronous optimization." When the time-varying rates of energy consumption and cost are large (e.g., energy consumption decreases but costs increase), the value of this term increases, the F-value rises, and the algorithm automatically adjusts the parameters to make the trends of both converge.

[0061] energy consumption E With the aforementioned operation and maintenance costs C Basic target weight k 1. k The determination of 2 includes:

[0062] Obtain suspended matter data from the water quality data, and classify the scene by combining it with a preset first threshold and a second threshold:

[0063] If the suspended matter data is less than the first threshold, the scenario is considered to be safe for water quality.

[0064] If the suspended matter data is greater than the first threshold and less than the second threshold, the scenario is considered to be at the critical water quality threshold.

[0065] If the suspended matter data exceeds the second threshold, the scenario is identified as a water quality warning.

[0066] The first threshold is less than the second threshold;

[0067] Obtain the pre-defined basic target weights based on the current scenario. k 1. k 2.

[0068] In this embodiment, adjustments are dynamically made based on real-time water quality data and equipment status. k 1. k The possible values ​​of 2 are shown in Table 1, for example:

[0069] Table 1. Parameter Comparison Table

[0070] Operating scenarios <![CDATA[ k 1 (Energy Consumption Weight) <![CDATA[ k 2 (Cost Weight) Water quality is safe (suspended solids <100mg / L) 0.7 0.3 Critical water quality (100 mg / L ≤ suspended solids < 400 mg / L) 0.5 0.5 Water quality warning (suspended solids ≥ 400 mg / L) 0.3 0.7

[0071] Nonlinear interaction term coefficient k3: ranges from 0.1 to 0.5, and is dynamically adjusted according to the equipment operating status: when the filter material clogging rate is <30%, k3=0.1; when the clogging rate is 30%-60%, k3=0.3; when the clogging rate is >60%, k3=0.5, thus strengthening the coupling constraint between energy consumption and cost.

[0072] The dynamic marginal effect coefficient k4 has a value range of 0.05-0.2, and is fixed at 0.15 in this embodiment. It is used to penalize the difference between the rate of change of energy consumption and cost, and to ensure that the two are optimized in sync.

[0073] and These are the time-varying rates of energy consumption and operation and maintenance costs, respectively, calculated using the finite difference method, as follows:

[0074] ;

[0075] Among them, △ t It lasts for 1 hour.

[0076] At the same time, in conjunction with aquaculture water quality standards and equipment operation safety requirements, three types of constraints are set:

[0077] Water quality constraints: After filtration, the suspended solids removal rate of the water is ≥95%, the ammonia nitrogen content is ≤0.2mg / L (≤0.15mg / L for Litopenaeus vannamei farming), the dissolved oxygen is ≥5mg / L, and the pH value is 7.5-8.5.

[0078] Equipment operating constraints: water pump operating pressure difference ≤ 0.2MPa, backwashing cycle ≥ 24 hours, daily operating time of disinfection equipment ≥ 2 hours, water pump speed adjustment range 1000-3000rpm.

[0079] Constraints of aquaculture scenarios: Differentiated thresholds are set according to the aquaculture species. For example, the ammonia nitrogen content threshold for grass carp farming is ≤0.3mg / L, and the backwashing cycle can be extended to 36 hours.

[0080] Energy consumption quota constraint: The total daily energy consumption shall not exceed the energy consumption quota of the aquaculture enterprise.

[0081] S5. The nonlinear coupling optimization model is optimized and solved using the improved NSGA-Ⅲ algorithm, and the filtering equipment is optimized and controlled based on the solution results;

[0082] The optimization solution of the nonlinear coupled optimization model using the improved NSGA-Ⅲ algorithm includes the following steps:

[0083] Using the water pump power, water pump running time, backwash water pump power, backwash time, disinfection equipment power, and disinfection equipment running time as core optimization variables, the corresponding feasible domain and adjustment rules are obtained.

[0084] Using a real-number encoding method, each individual is defined as a set of the core optimization variables. An initial population is generated by combining historical running data with random generation, and the fitness value of each individual in the initial population is calculated. F ;

[0085] Based on the initial population and the preset maximum number of iterations, the optimal solution set is generated through iterative solutions using adaptive crossover mutation, non-dominated sorting, and crowding calculation.

[0086] In this embodiment, six types of parameters—pump power, pump runtime, backwash pump power, backwash duration, disinfection equipment power, and disinfection equipment runtime—are used as the core optimization variables of the algorithm. The physical meaning, feasible region, and adjustment rules of each variable are clearly defined to ensure that the variables conform to the actual constraints of equipment operation, as shown in Table 2.

[0087] Table 2. Variable Comparison Table

[0088] Optimize variables Physical meaning Feasible domain range Adjusting rules Water pump power Real-time power of main water pump (kW) 30%-100% of rated power (e.g., 5-15kW) Continuous adjustment via frequency converter, in steps of 0.1kW. Water pump running time Daily operating time of the main water pump (h) 20-24h It can be dynamically adjusted according to water quality fluctuations, and can guarantee continuous operation for at least 20 hours to maintain water quality. Backwash water pump power Real-time power of backwash pump (kW) 50%-100% of the rated power (e.g., 3-6kW) Fixed power adjustment, with three power levels: high, medium, and low. Backwash duration Duration of a single backwash (min) 10-30 min Adjust the backwashing time according to the degree of filter media clogging; the higher the clogging rate, the longer the backwashing time. Disinfection equipment power Real-time power of disinfection equipment (kW) 40%-100% of the rated power (e.g., 2-5kW) Adjustments are made according to the degree of water pollution; the higher the ammonia nitrogen content, the higher the power. Disinfection equipment running time Daily operating time of disinfection equipment (h) 2-6h It can be run in segments, such as two 1.5-hour runs, to ensure disinfection effectiveness.

[0089] Using real-number encoding, each individual corresponds to a set of 6-dimensional optimization parameters, with the encoding format as follows:

[0090] X=[ P pump , t pump , P backwash , t backwash , P disinfection , t disinfection ];

[0091] For example, X=[12.5, 23, 5.2, 20, 3.8, 3.5] means that the main water pump has a power of 12.5kW and runs for 23 hours a day, the backwash pump has a power of 5.2kW and a single flushing time of 20 minutes, and the disinfection equipment has a power of 3.8kW and runs for 3.5 hours a day.

[0092] The population size is set to N=120 (to ensure solution diversity, the population size is appropriately increased as the dimensionality of the dependent variable increases). A hybrid initialization method of "historical running data + random generation" is adopted:

[0093] We selected 30 parameter combinations with the best overall performance (low energy consumption, controllable cost, and water quality compliance) from the equipment operation data of the past 3 months, and used them as 70% of the initial population.

[0094] The remaining 30% of individuals are generated through random sampling to ensure coverage of the feasible domain of all variables and avoid the algorithm getting trapped in local optima.

[0095] In addition to the existing water quality constraints and equipment operation constraints, new specific constraints for the optimization variables are added to ensure that the solution results meet the requirements for safe equipment operation.

[0096] Power constraints: The real-time power of all equipment must not exceed the rated power, and the minimum power must meet the equipment start-up requirements (e.g., the minimum power of water pumps should not be less than 30% of the rated power to avoid motor stall).

[0097] Duration constraints: The main water pump shall run for no less than 20 hours per day to ensure the water circulation efficiency of the aquaculture pond; the backwashing cycle shall not be less than 24 hours to avoid frequent backwashing that would lead to increased wear and tear on the filter media.

[0098] Energy consumption related constraints: Total daily energy consumption

[0099] E = P pump × t pump + P backwash × t backwash + P disinfection × t disinfection ;

[0100] Total energy consumption must not exceed the daily energy consumption quota of the aquaculture enterprise.

[0101] The core optimization variables (6 types of parameters) of the algorithm are directly substituted into the nonlinear coupled objective function to calculate the fitness value of each individual, that is:

[0102] .

[0103] For each individual, the parameter combination is constrained and verified. If the power, duration, or water quality constraints are violated, the fitness value is corrected using a penalty function method.

[0104] ;

[0105] in, To determine the degree of violation of the i-th constraint, The penalty coefficient (when power constraints are applied) =8, corresponding to the duration constraint =5, corresponding to water quality constraints =10), ensuring that individuals who violate the constraints are at a disadvantage in the ranking.

[0106] Following the core logic of adaptive crossover and mutation, non-dominated sorting, and crowding calculation, the algorithm parameters are adjusted for the 6-dimensional optimization variables:

[0107] Adaptive crossover mutation probability: dynamically adjusted based on the standard deviation σ of the population fitness.

[0108] when σ <0.1 (population convergence), crossover probability p c =0.85, mutation probability p m =0.25, enhancing population diversity;

[0109] when σ ≥0.1 (population dispersed), crossover probability p c =0.65, mutation probability p m =0.15, to speed up the convergence.

[0110] Iteration termination condition: Set the maximum number of iterations G. max =250 (Increase the number of iterations as the dependent variable dimension increases), and set an early termination condition: terminate the iteration when the change in the optimal fitness value ΔF < 0.01 for 25 consecutive generations.

[0111] Individuals from the first Pareto front layer are extracted from the final population to form the optimal solution set. The set size is controlled at 25-35 individuals, covering parameter combinations with different priorities (such as energy saving priority, cost priority, and water quality priority).

[0112] Based on the water quality data, the scenario is classified, and the first solution is selected from the set of optimal solutions based on the classification results;

[0113] In this embodiment, a suitable solution is selected from the optimal solution set based on real-time water quality data (such as suspended solids concentration and ammonia nitrogen content). For example:

[0114] When a water quality warning is issued (suspended solids ≥ 400 mg / L): P should be selected as the first choice. pump Higher, t backwash A longer-term plan ensures that water quality meets standards quickly;

[0115] When water quality is safe (suspended solids <100mg / L): P is the preferred choice. pump Lower, t disinfection A shorter solution reduces energy consumption and costs;

[0116] Execution parameters are generated based on the first solution and sent to the PLC controller, which then optimizes the control of the filtration device.

[0117] According to another aspect of the invention, Figure 2 This is a schematic diagram illustrating a system for collaboratively optimizing the energy consumption and operation and maintenance costs of a filtration device according to an embodiment of the present invention.

[0118] Embodiment 2 of the present invention is as follows:

[0119] A system 1 for collaboratively optimizing the energy consumption and operation and maintenance costs of a filtration device includes:

[0120] Data acquisition module 2 collects equipment operation data of the filtration equipment, synchronously collects water quality data using online sensors installed in the aquaculture pond, and connects to the enterprise management system to obtain cost data;

[0121] Model building module 3 constructs a functional representation of energy consumption based on the pump power, pump running time, backwash pump power, backwash time, disinfection equipment power, and disinfection equipment running time in the equipment operation data.

[0122] Based on the energy consumption, electricity price per unit, maintenance hours, hourly labor cost, filter media replacement frequency, filter media purchase cost, equipment purchase cost, depreciation rate, downtime due to failure, and the aquaculture loss per unit time due to downtime due to failure, a functional representation of operation and maintenance cost is constructed.

[0123] Weighting coefficients are assigned to the energy consumption and the operation and maintenance cost, and nonlinear interaction terms and dynamic marginal effect terms are introduced to construct a nonlinear coupled optimization model of the energy consumption and the operation and maintenance cost, and constraints are constructed.

[0124] The weighting coefficients for energy consumption and operation and maintenance costs are determined based on the water quality data.

[0125] Model solving module 4 optimizes and solves the nonlinear coupling optimization model using the improved NSGA-Ⅲ algorithm, and optimizes and controls the filtering device based on the solution results.

[0126] Based on the cooperation of the data acquisition module 2, model building module 3 and model solving module 4, the steps in the collaborative optimization method of energy consumption and operation and maintenance cost of a filtration device in the above embodiment 1 are realized.

[0127] This invention discloses a method and system for the coordinated optimization of energy consumption and operation and maintenance costs of filtration devices. Through an integrated design encompassing "full-dimensional data collection, precise dual-objective modeling, coupled dynamic optimization, efficient algorithm solution, and intelligent closed-loop control," it achieves deep coordinated optimization of energy consumption and operation and maintenance costs, resulting in significant overall benefits. This invention breaks away from the limitations of traditional single-optimization methods. It first integrates three core data categories—equipment operation, water quality, and cost—to lay a comprehensive data foundation for coordinated optimization. Then, it precisely quantifies energy consumption and operation and maintenance costs through explicit functional expressions, clearly depicting the logical relationship between the two. The core lies in constructing a nonlinear coupled model based on dynamically adjusted weights from water quality data, introducing nonlinear interaction terms and dynamic marginal effect terms to adapt to different water quality scenarios and equipment operating states, thus resolving the coupling contradiction between energy consumption and operation and maintenance costs. By efficiently solving for the optimal solution using an improved NSGA-Ⅲ algorithm and combining it with a PLC controller for automated control, it avoids both insufficient filtration efficiency and a surge in failures caused by simply pursuing low energy consumption, and eliminates energy waste caused by excessive pursuit of convenient operation and maintenance, achieving a two-way reduction in both. This invention is adapted to the needs of large-scale and intensive aquaculture, significantly improving the economy and sustainability of aquaculture systems, providing key technical support for the industry's efficient, low-carbon, and intelligent transformation, and possessing both practicality and promotional value.

[0128] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for synergistic optimization of energy consumption and operation and maintenance costs of a filtration device, characterized in that, Including the following steps: S1. Collect equipment operation data of the filtration equipment, collect water quality data synchronously using online sensors installed in the aquaculture pond, and connect to the enterprise management system to obtain cost data; S2. Construct a function representation of energy consumption based on the pump power, pump running time, backwash pump power, backwash time, disinfection equipment power, and disinfection equipment running time in the equipment operation data; S3. Based on the energy consumption, electricity price, maintenance hours, hourly labor cost, filter media replacement frequency, filter media purchase cost, equipment purchase cost, depreciation rate, downtime due to failure, and the aquaculture loss per unit time due to downtime due to failure, construct a functional representation of the operation and maintenance cost; S4. Assign weight coefficients to the energy consumption and the operation and maintenance cost, and introduce nonlinear interaction terms and dynamic marginal effect terms to construct a nonlinear coupled optimization model of the energy consumption and the operation and maintenance cost, and construct constraints. The weighting coefficients for energy consumption and operation and maintenance costs are determined based on the water quality data. S5. The nonlinear coupling optimization model is optimized and solved using the improved NSGA-Ⅲ algorithm, and the filtering equipment is optimized and controlled based on the solution results.

2. The method for synergistic optimization of energy consumption and operation and maintenance costs of a filtration device according to claim 1, characterized in that, energy consumption E The function is specifically represented as follows: E = P pump × t pump + P backwash × t backwash + P disinfection × t disinfection ； in, P pump Indicates the power of the water pump. t pump Indicates the operating time of the water pump. P backwash Indicates the power of the backwash water pump. t backwash Indicates the backwash duration. P disinfection Indicates the power of the disinfection equipment. t disinfection This indicates the operating time of the disinfection equipment.

3. The method for synergistic optimization of energy consumption and operation and maintenance costs of a filtration device according to claim 1, characterized in that, The operation and maintenance costs C The function is specifically represented as follows: C = C electrity + C labor + C filter + C depreciation + C failure ； C electrity = E × U ； C labor = T × R ； C filter = N × P filter ； C depreciation = P depreciation × r ； C failure = L × t failure ； in, C electrity Indicates electricity cost. U This indicates the unit price of electricity. E Indicates energy consumption. C labor Indicates maintenance costs, T Indicates maintenance time. R This represents hourly labor costs. C filter Indicates the cost of filter media. N Indicates the number of times the filter media has been replaced. P filter This indicates the cost of a single filter material purchase. C depreciation This represents the cost of equipment depreciation. P depreciation This indicates the cost of purchasing the equipment. r Indicates the depreciation rate. C failure Indicates the cost of failure and damage. L Indicates the duration of downtime due to a fault. t failure This represents the aquaculture loss per unit of time caused by the downtime due to a malfunction.

4. The method for synergistic optimization of energy consumption and operation and maintenance costs of a filtration device according to claim 1, characterized in that, energy consumption E With the aforementioned operation and maintenance costs C The nonlinear coupling optimization model is expressed as: ； in, F Indicates the fitness value. k 1. k 2 represents the energy consumption respectively. E With the aforementioned operation and maintenance costs C The basic target weights are obtained based on fuzzy inference from the water quality data. k 1+ k 2=1, k 3 indicates the coefficient of the nonlinear interaction term, which is dynamically adjusted according to the equipment's operating status. k 4 represents the coefficient of the dynamic marginal effect term.

5. The method for synergistic optimization of energy consumption and operation and maintenance costs of a filtration device according to claim 4, characterized in that, energy consumption E With the aforementioned operation and maintenance costs C Basic target weight k 1. k The determination of 2 includes: Obtain suspended matter data from the water quality data, and classify the scene by combining it with a preset first threshold and a second threshold: If the suspended matter data is less than the first threshold, the scenario is considered to be safe for water quality. If the suspended matter data is greater than the first threshold and less than the second threshold, the scenario is considered to be at the critical water quality threshold. If the suspended matter data exceeds the second threshold, the scenario is identified as a water quality warning. The first threshold is less than the second threshold; Obtain the pre-defined basic target weights based on the current scenario. k 1. k 2.

6. The method for synergistic optimization of energy consumption and operation and maintenance costs of a filtration device according to claim 4, characterized in that, The optimization solution of the nonlinear coupled optimization model using the improved NSGA-Ⅲ algorithm includes the following steps: Using the water pump power, water pump running time, backwash water pump power, backwash time, disinfection equipment power, and disinfection equipment running time as core optimization variables, the corresponding feasible domain and adjustment rules are obtained. Using a real-number encoding method, each individual is defined as a set of the core optimization variables. An initial population is generated by combining historical running data with random generation, and the fitness value of each individual in the initial population is calculated. F ; Based on the initial population and the preset maximum number of iterations, the optimal solution set is generated through iterative solutions using adaptive crossover mutation, non-dominated sorting, and crowding calculation.

7. The method for synergistic optimization of energy consumption and operation and maintenance costs of a filtration device according to claim 6, characterized in that, Optimizing and controlling the filtration equipment based on the solution results includes: Based on the water quality data, the scenario is classified, and the first solution is selected from the set of optimal solutions according to the classification results; Execution parameters are generated based on the first solution and sent to the PLC controller, which then optimizes the control of the filtration device.

8. A system for collaboratively optimizing the energy consumption and operation and maintenance costs of a filtration device, characterized in that, include: The data acquisition module collects equipment operation data from the filtration equipment, synchronously collects water quality data using online sensors installed in the aquaculture pond, and connects to the enterprise management system to obtain cost data. The model building module constructs a functional representation of energy consumption based on the pump power, pump runtime, backwash pump power, backwash duration, disinfection equipment power, and disinfection equipment runtime in the equipment operation data. Based on the energy consumption, electricity price per unit, maintenance hours, hourly labor cost, filter media replacement frequency, filter media purchase cost, equipment purchase cost, depreciation rate, downtime due to failure, and the aquaculture loss per unit time due to downtime due to failure, a functional representation of operation and maintenance cost is constructed. Weighting coefficients are assigned to the energy consumption and the operation and maintenance cost, and nonlinear interaction terms and dynamic marginal effect terms are introduced to construct a nonlinear coupled optimization model of the energy consumption and the operation and maintenance cost, and constraints are constructed. The weighting coefficients for energy consumption and operation and maintenance costs are determined based on the water quality data. The model solving module optimizes and solves the nonlinear coupling optimization model using the improved NSGA-Ⅲ algorithm, and optimizes and controls the filtering device based on the solution results.

9. A system for collaborative optimization of energy consumption and operation and maintenance costs of a filtration device according to claim 8, characterized in that, energy consumption E With the aforementioned operation and maintenance costs C The nonlinear coupling optimization model is expressed as: ； in, F Indicates the fitness value. k 1. k 2 represents the energy consumption respectively. E With the aforementioned operation and maintenance costs C The basic target weights are obtained based on fuzzy inference from the water quality data. k 1+ k 2=1, k 3 indicates the coefficient of the nonlinear interaction term, which is dynamically adjusted according to the equipment's operating status. k 4 represents the coefficient of the dynamic marginal effect term.

10. The system for collaborative optimization of energy consumption and operation and maintenance costs of a filtration device according to claim 8, characterized in that, The optimization solution of the nonlinear coupled optimization model using the improved NSGA-Ⅲ algorithm includes the following steps: Using the water pump power, water pump running time, backwash water pump power, backwash time, disinfection equipment power, and disinfection equipment running time as core optimization variables, the corresponding feasible domain and adjustment rules are obtained. Using a real-number encoding method, each individual is defined as a set of the core optimization variables. An initial population is generated by combining historical running data with random generation, and the fitness value of each individual in the initial population is calculated. F ; Based on the initial population and the preset maximum number of iterations, the optimal solution set is generated through iterative solutions using adaptive crossover mutation, non-dominated sorting, and crowding calculation.

Images