A method, system, device and medium for constructing an activity-cost conversion model

Abstract

This invention provides a method, system, equipment, and medium for constructing an activity-cost conversion model. The method includes obtaining activity feature vectors and comprehensive operating costs; performing feature engineering based on the activity feature vectors and comprehensive operating costs to obtain target feature vector pairs; and constructing an activity-cost conversion model based on the target feature vector pairs and machine learning algorithms. The activity-cost conversion model constructed using this method establishes a real-time, dynamic mathematical correlation between key technical parameters characterizing sludge health and treatment capacity and the final comprehensive operating cost per ton of wastewater through a data-driven approach. This makes cost control no longer an independent task parallel to or even conflicting with process optimization, but rather enables the measurement and optimization of technical activities using economic cost as an intuitive benchmark. This drives wastewater treatment plants to intelligently and accurately move towards a global balance between "stable compliance" and "optimal cost."

Description

Technical Field

[0001] This invention relates to the field of wastewater treatment technology, and in particular to a method, system, equipment, and medium for constructing an activity-cost conversion model. Background Technology

[0002] In the daily operation of wastewater treatment plants, operators face a fundamental contradiction: to achieve better effluent quality and stronger shock resistance, they typically tend to maintain high sludge activity (e.g., maintaining high DO and MLSS), but this often leads to increased aeration energy consumption and sludge treatment costs; conversely, excessively reducing activity to save costs may result in effluent exceeding standards. Traditionally, activity control and cost control have been two separate decision-making systems, relying on experience to make a difficult trade-off between "effectiveness" and "cost." However, existing technologies and management practices have shortcomings. (1) Technical indicators and economic indicators are seriously disconnected: The current operation mainly monitors technical indicators such as DO, MLSS, and SOUR, as well as economic data such as electricity consumption and drug costs. However, these two types of data are usually recorded and analyzed separately. Operators cannot intuitively know "what is the cost per ton of water when SOUR is maintained at X mgO2 / gMLSS·h today", nor can they predict "how much the electricity cost will increase if MLSS is increased by 100 mg / L". Decision-making relies on vague experience and lacks data support.

[0003] (2) Cost accounting is lagging and crude: Traditional cost accounting is ex post and periodic (such as monthly accounting), which cannot reflect the immediate economic effects of process adjustments in real time. At the same time, cost allocation is crude and it is difficult to accurately link total power consumption and drug consumption to specific activity control actions, resulting in the inability to conduct accurate benefit assessment and root cause analysis.

[0004] (3) Lack of dynamic cost prediction and optimization tools: Most existing optimization models focus on process simulation or single cost constraints, lacking a dedicated model that can dynamically learn and establish a nonlinear mapping relationship between "activity parameters" and "comprehensive cost". Therefore, the system cannot predict the optimal cost point under different activity levels in changing environments (such as different influent loads and temperatures), nor can it actively find the optimal process path to reach that point.

[0005] Therefore, how to solve the disconnect between process control and economic costs is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0006] To address the aforementioned issues, this invention provides a method, system, equipment, and medium for constructing an activity-cost conversion model. By constructing this model, key technical parameters characterizing sludge health and treatment capacity (such as SOUR, MLSS, and SRT) are mathematically correlated in real-time with the final comprehensive operating cost per ton of wastewater through a data-driven approach. This ensures that cost control is no longer an independent task parallel to or even conflicting with process optimization, but rather is intrinsically integrated into every process adjustment decision. The activity-cost conversion model enables the measurement and optimization of technical activities using economic cost as a direct benchmark, thereby driving wastewater treatment plants towards a global balance between "stable compliance" and "optimal cost."

[0007] The first objective of this invention is to provide a method for constructing an activity-cost conversion model; The technical solution provided by this invention is as follows: A method for constructing an activity-cost conversion model includes the following steps: Obtain the active feature vector and the overall operating cost; Feature engineering is performed based on the active feature vector and the comprehensive operating cost to obtain target feature vector pairs; An activity-cost conversion model is constructed based on the target feature vector pairs and machine learning algorithms.

[0008] Preferably, obtaining the active feature vector specifically includes: Historical and current operating data are collected and stored in real time through IoT gateways; Extract the time series of characteristic parameters that are directly or strongly correlated with sludge activity from the historical operating data and the current operating data to obtain the activity feature vector.

[0009] Preferably, obtaining the comprehensive operating cost specifically includes: The comprehensive operating cost is calculated based on the operating data concurrent with the activity parameters, wherein the specific formula for calculating the comprehensive operating cost is as follows: ; in, This represents the overall operating cost; This represents the electricity consumption cost directly related to biochemical treatment; This indicates the cost of pharmaceutical consumption; This indicates the amount of wastewater treated within the corresponding time period.

[0010] Preferably, the step of constructing feature engineering based on the active feature vector and the comprehensive operating cost to obtain target feature vector pairs specifically includes: The active feature vector and the comprehensive operating cost are aligned according to a time window to obtain the target active feature vector and the target comprehensive operating cost; Feature derivation and filtering are performed based on the target activity feature vector and the target comprehensive operating cost to obtain target feature vector pairs.

[0011] Preferably, the step of constructing an activity-cost conversion model based on the target feature vector pair and a machine learning algorithm specifically includes: Build an initial model based on machine learning algorithms; The target feature vector is used to train the initial model to obtain the activity-cost conversion model.

[0012] Preferably, after constructing the activity-cost conversion model based on the target feature vector pair and the machine learning algorithm, the method further includes: Cost prediction and economic evaluation are performed using the activity-cost conversion model.

[0013] Preferably, the step of performing cost prediction and economic evaluation using the activity-cost conversion model specifically includes: Collect current activity characteristic data and input it into the activity-cost conversion model to obtain the predicted cost per ton of water under the current operating conditions; Calculate the average activity characteristics of water that is stably compliant with standards and within the target time period, and calculate the reference cost corresponding to the average activity characteristics based on the activity-cost conversion model. An economic assessment is conducted based on the predicted cost per ton of water and the reference cost.

[0014] The second objective of this invention is to provide a system for constructing an activity-cost conversion model; The technical solution provided by this invention is as follows: A system for constructing an activity-cost conversion model includes: a first acquisition module, a second acquisition module, and a construction module; The first acquisition module is used to acquire the active feature vector and the overall operating cost; The second acquisition module is used to perform feature engineering construction based on the active feature vector and the comprehensive operating cost to obtain target feature vector pairs; The construction module is used to construct an activity-cost conversion model based on the target feature vector pair and the machine learning algorithm.

[0015] The third objective of this invention is to provide an electronic device; The technical solution provided by this invention is as follows: An electronic device, comprising: At least one processor; and A memory communicatively connected to the at least one processor, the memory storing a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the method steps of any one of the methods for constructing the activity-cost conversion model.

[0016] A fourth objective of this invention is to provide a computer-readable storage medium; The technical solution provided by this invention is as follows: A computer-readable storage medium for storing a computer program for causing a computer to perform the steps of any one of the methods for constructing an activity-cost conversion model.

[0017] Compared with existing technologies, this invention provides a method for constructing an activity-cost conversion model, including obtaining activity feature vectors and comprehensive operating costs; performing feature engineering based on the activity feature vectors and comprehensive operating costs to obtain target feature vector pairs; and constructing an activity-cost conversion model based on the target feature vector pairs and machine learning algorithms. The activity-cost conversion model constructed by this method establishes a real-time, dynamic mathematical correlation between key technical parameters characterizing sludge health and treatment capacity (such as SOUR, MLSS, and SRT) and the final comprehensive operating cost per ton of wastewater through a data-driven approach. This makes cost control no longer an independent task parallel to or even conflicting with process optimization, but rather intrinsic to every process adjustment decision. The activity-cost conversion model enables the measurement and optimization of technical activities using economic costs as an intuitive benchmark, thereby driving wastewater treatment plants to intelligently and accurately move towards a global balance point of "stable compliance" and "optimal cost."

[0018] The present invention also provides a system for constructing an activity-cost conversion model. Since the system and the method for constructing the activity-cost conversion model solve the same technical problem and belong to the same technical concept, they should have the same beneficial effects, and will not be described in detail here. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A flowchart illustrating a method for constructing an activity-cost conversion model provided in this application embodiment; Figure 2A schematic diagram of the structure of a system for constructing an activity-cost conversion model provided in this application embodiment; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0021] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] like Figure 1 As shown, an embodiment of the present invention provides a method for constructing an activity-cost conversion model, comprising the following steps: S1. Obtain the active feature vector and overall operating cost; S2. Perform feature engineering based on the active feature vector and the comprehensive operating cost to obtain target feature vector pairs; S3. Construct an activity-cost conversion model based on the target feature vector pair and machine learning algorithm.

[0023] In steps S1 to S3, firstly, the activity feature vectors of microorganisms are acquired, and the comprehensive operating cost of the system is calculated in detail. Then, based on the obtained activity feature vectors and comprehensive operating cost data, in-depth feature engineering is carried out. Through feature selection, dimensionality reduction, and combinatorial optimization, target feature vector pairs with higher discriminative power are generated. Finally, using the above target feature vector pairs as input, combined with appropriate machine learning algorithms, a conversion model that accurately reflects the dynamic relationship between microbial activity and operating cost is constructed to achieve efficient cost prediction and optimization. The activity-cost conversion model constructed by this method, namely the microbial activity-cost dynamic conversion model, establishes a real-time, dynamic mathematical correlation between key technical parameters characterizing sludge health and treatment capacity (such as SOUR, MLSS, and SRT) and the final comprehensive operating cost per ton of water through a data-driven approach. This makes cost control no longer an independent task parallel to or even conflicting with process optimization, but rather endogenous to every process adjustment decision. The microbial activity-cost dynamic conversion model enables the measurement and optimization of technical activities using economic costs as an intuitive benchmark, thereby driving wastewater treatment plants to intelligently and accurately move towards the global balance point of "stable compliance" and "optimal cost".

[0024] Preferably, obtaining the active feature vector specifically includes: Historical and current operating data are collected and stored in real time through IoT gateways; Extract the time series of characteristic parameters that are directly or strongly correlated with sludge activity from the historical operating data and the current operating data to obtain the activity feature vector.

[0025] In practical applications, historical and current operational data are collected and stored in real time through IoT gateways. The focus is on extracting time-series characteristic parameters that are directly or strongly correlated with sludge activity, forming an activity feature vector A_t=[a1_t,a2_t,...,an_t], where key features include at least: a1:SOUR (specific oxygen consumption rate, mgO2 / gMLSS·h), a core metabolic activity indicator.

[0026] a2: MLSS (mixed liquor suspended solids concentration, mg / L), a basic indicator of biomass.

[0027] a3:SRT (sludge age, days) is a key parameter reflecting the composition and stability of the microbial community.

[0028] a4:DO (dissolved oxygen concentration, mg / L), an environmental indicator of aerobic activity.

[0029] a5:F / M (food-to-microbe ratio) reflects the balance between organic load and biomass.

[0030] Other optional features include OUR, nitrification rate, denitrification rate, and SVI (sludge volume index).

[0031] Preferably, obtaining the comprehensive operating cost specifically includes: The comprehensive operating cost is calculated based on the operating data concurrent with the activity parameters, wherein the specific formula for calculating the comprehensive operating cost is as follows: ; in, This represents the overall operating cost; This represents the electricity consumption cost directly related to biochemical treatment; This indicates the cost of pharmaceutical consumption; This indicates the amount of wastewater treated within the corresponding time period.

[0032] In practical applications, a refined comprehensive operating cost per ton of water, C_t (yuan / ton), is calculated based on operational data concurrent with the activity parameters. This calculation differs from traditional financial accounting and requires precise attribution. ; Here, E_t represents the electricity cost (in yuan) directly related to the biochemical treatment, focusing on the electricity consumption of the aeration system, and may also include the electricity consumption of pumps strongly related to activity regulation, such as booster and reflux pumps. This data is obtained through separate metering or model estimation based on equipment power and operating time.

[0033] M_t represents the cost of reagent consumption (in yuan), which mainly includes the cost of adding carbon source (such as sodium acetate) and the cost of adding phosphorus removal agent (such as PAC).

[0034] Q_t represents the wastewater treatment volume (tons) within the corresponding time period.

[0035] It is important to note that it is necessary to ensure that the comprehensive operating cost C_t of each ton of water can accurately correspond to the activity feature vector A_t of the same period.

[0036] Preferably, the step of constructing feature engineering based on the active feature vector and the comprehensive operating cost to obtain target feature vector pairs specifically includes: The active feature vector and the comprehensive operating cost are aligned according to a time window to obtain the target active feature vector and the target comprehensive operating cost; Feature derivation and filtering are performed based on the target activity feature vector and the target comprehensive operating cost to obtain target feature vector pairs.

[0037] In practical applications, the active feature vector and the comprehensive operating cost are aligned and cleaned. Specifically, the active feature vector A_t and the comprehensive operating cost C_t are strictly aligned according to time windows (such as by day or by shift). Then, invalid or abnormal data segments caused by instrument failure or production maintenance are removed to obtain the target active feature vector and the target comprehensive operating cost.

[0038] Feature derivation and screening are performed based on the target activity feature vector and the target comprehensive operating cost. A feature importance evaluation method (such as feature importance scoring based on XGBoost) is used to select the core feature subset that contributes the most to the prediction of the comprehensive operating cost C_t from the original and derived features, forming the final feature vector pair [X_t, C_t]. In this embodiment, Pearson, Spearman, and Kendall correlation calculation methods are used, considering the linear and nonlinear relationships between the feature sequence and the target cost sequence, respectively. The calculation methods are as follows: (1) The Pearson correlation coefficient requires that the data follow a normal distribution and that there is a linear relationship between the variables. It is sensitive to outliers. The formula for the Pearson correlation coefficient is: ; Where x i and y i These are the observed values ​​of two variables. and These are the sample means of the observed values ​​of the two variables, and n is the sample size.

[0039] The Pearson correlation coefficient measures the degree of linear correlation between two continuous variables, and its value ranges from -1 to 1. r=1 indicates a perfectly positive linear correlation; r=-1 indicates a perfectly negative linear correlation; r=0 indicates no linear correlation.

[0040] (2) Spearman's rank correlation coefficient formula: ; Where d i It is each pair of observations (x) i The difference in rank between ,yi, i.e., d i =rank(x i )-rank(y i ); n is the sample size. If there are samples with the same rank, a more complex correction formula is required.

[0041] The Spearman correlation coefficient is a nonparametric measure that assesses the strength and direction (not necessarily linear) of a monotonic relationship between two variables. It is based on the rank of the data rather than the original values, and its value ranges from -1 to 1. It is suitable for ordinal scale data or nonlinear monotonic relationships and is insensitive to outliers.

[0042] The commonly used formula for Kendall's rank correlation coefficient (Kendallilstau) (Kendall's t_b): ; Where: C is the number of concordant pairs: for any two observations (x... i ,y i ) and (x j ,y j If x i >x j And y i >y j , or x i <x j And y i <y j If they are a pair, then they are called a consistent pair.

[0043] D is the number of discordant pairs: if x i >x j And y i <y j, or x i <x j And y i >y j If they are not a pair, then they are called inconsistent pairs.

[0044] T x T is the logarithm that is equal only on x; y For logarithms that are equal only on y (i.e., handling cases with the same rank).

[0045] If there are no identical ranks, the formula simplifies to: .

[0046] Kendall's rank correlation coefficient is also a nonparametric measure that assesses the consistency between two ordinal variables. It is based on the rank relationship of pairwise comparisons of observations, and its value ranges from -1 to 1. Compared to Spearman's coefficient, Kendall's T is more robust to smaller sample sizes and provides a more intuitive interpretation (i.e., the proportion of differences between consistent and inconsistent pairs). It is commonly used to analyze ordinal or non-normally distributed data.

[0047] Finally, the average of the three methods is taken, and those with an absolute value of the average greater than or equal to 0.4 are considered to be valid features, i.e., feature vector pairs [X_t, C_t].

[0048] In addition to using the original activity parameters, this embodiment can also perform feature engineering to improve the model's expressive power, for example: Create interactive features such as SOUR*MLSS (reflecting total oxygen consumption potential) and F / M*SRT.

[0049] Create statistical features such as the mean, variance, and trend slope of the activity parameter over the past 24 hours.

[0050] The processed feature vectors [X_t, C_t] are then divided into training, validation, and test sets. Here, F / M*SRT represents the total amount of substrate removed per unit of microbial biomass within the sludge age (SRT). It is also the reciprocal of the apparent yield coefficient Yobs, where Yobs is the observed sludge yield (the ratio of daily sludge discharge to substrate removal). Therefore, this product directly reflects the balance between the system's substrate conversion efficiency and sludge production.

[0051] Preferably, the step of constructing an activity-cost conversion model based on the target feature vector pair and a machine learning algorithm specifically includes: Build an initial model based on machine learning algorithms; The target feature vector is used to train the initial model to obtain the activity-cost conversion model.

[0052] In practical applications, a machine learning algorithm suitable for regression problems is selected to construct an initial model M. The initial form of the initial model M can be expressed as: C_pred=M(X;θ), where θ is the parameter to be trained in the model, and X represents the input feature sequence left after thresholding above.

[0053] Preferred machine learning algorithms include, but are not limited to: Gradient boosting decision trees (such as XGBoost, LightGBM): which can effectively handle nonlinear relationships and feature interactions, are insensitive to missing values, and have relatively good interpretability; and neural networks (such as feedforward neural networks): which have strong fitting capabilities for extremely complex nonlinear mappings.

[0054] The target feature vectors used as the training set are used to train the initial model M, and the true cost C_train corresponding to the target feature vectors used as the training set is used as the output target. Use the target feature vectors, which serve as the validation set, to perform hyperparameter tuning (such as learning rate, tree depth, regularization parameters, etc.) to prevent overfitting.

[0055] The final prediction accuracy of the model is evaluated using the target feature vectors as the test set to ensure its generalization ability. The trained, accurate initial model M, along with its required feature preprocessing pipeline (such as a normalizer), is packaged and deployed to an online prediction service to obtain an activity-cost conversion model. This model abandons traditional static and linear cost estimation methods and, for the first time, applies machine learning regression algorithms (such as XGBoost and neural networks) to establish a high-dimensional, nonlinear, and dynamic quantitative relationship between key sludge activity parameters (SOUR, MLSS, SRT, etc.) and the comprehensive operating cost per ton of water. This model can automatically learn complex cost-driving patterns from historical data, achieving a precise "translation" from technical status to economic performance.

[0056] In this embodiment, the loss function used for training is typically the mean squared error (MSE) or the mean absolute percentage error (MAPE) to minimize the gap between the predicted cost C_pred and the true cost C_true.

[0057] To ensure the activity-cost conversion model can adapt to changes and maintain long-term effectiveness, continuous model evolution and knowledge accumulation are necessary. This involves recording all decisions predicted based on the activity-cost conversion model and their actual execution results, including the new activity parameter X_new and the actual calculated true cost C_true_new. (X_new, C_true_new) constitute new high-quality sample pairs.

[0058] Add new sample pairs to the training dataset regularly (e.g., weekly or monthly).

[0059] Triggering the incremental learning or full retraining process of the activity-cost conversion model updates the model parameters θ. This enables the activity-cost conversion model to dynamically track and learn the evolution of the "activity-cost" relationship caused by seasonal changes, changes in influent water quality, decline in equipment efficiency, fluctuations in energy prices, etc.

[0060] Preferably, after constructing the activity-cost conversion model based on the target feature vector pair and the machine learning algorithm, the method further includes: Cost prediction and economic evaluation are performed using the activity-cost conversion model.

[0061] In practical applications, this activity-cost conversion model is used for cost prediction and economic evaluation. Based on the dynamic relationship between changes in microbial activity and costs, the model can accurately predict production costs under different conditions and further evaluate economic benefits and feasibility, providing data support for decision-making.

[0062] Preferably, the step of performing cost prediction and economic evaluation using the activity-cost conversion model specifically includes: Collect current activity characteristic data and input it into the activity-cost conversion model to obtain the predicted cost per ton of water under the current operating conditions; Calculate the average activity characteristics of water that is stably compliant with standards and within the target time period, and calculate the reference cost corresponding to the average activity characteristics based on the activity-cost conversion model. An economic assessment is conducted based on the predicted cost per ton of water and the reference cost.

[0063] In practical applications, the current activity characteristic data X_current is collected in real time and input into the deployed activity-cost conversion model to calculate the predicted cost per ton of water C_pred_current under the current operating conditions. This predicted value can be displayed on the operation dashboard in real time as a "cost dashboard".

[0064] From historical data, the "golden operation" period, characterized by stable water quality and the lowest total cost, is selected as the target period. The average activity characteristic X_optimal for this period is calculated, and the corresponding reference optimal cost C_optimal is calculated using the activity-cost conversion model. The current predicted cost C_pred_current is then compared with the reference optimal cost C_optimal. If (predicted cost per ton of water C_pred_current - reference optimal cost C_optimal) / reference optimal cost C_optimal > δ (δ is a set threshold, such as 10%), an economic warning will be triggered, indicating that the current operating mode deviates from the economic optimal state.

[0065] By utilizing the activity-cost conversion model, the system can predict the current "expected cost per ton of water" online and instantaneously based on real-time collected activity parameters, transforming the originally lagging and periodic financial accounting into a real-time monitoring "cost dashboard." Simultaneously, by comparing with the historical best cost benchmark, the system can dynamically assess the economic efficiency of the current operating mode and issue immediate warnings of economic deviations.

[0066] When process adjustments are required, the system activates the decision simulation engine: under the constraint of ensuring water quality meets standards, it automatically searches or optimizes within a reasonable range of activity parameters to find the activity parameter setting combination X_recommended that minimizes the predicted cost C_pred of the activity-cost conversion model, and recommends this combination to the operator or issues it to the underlying intelligent agent as the setting target. When increased processing capacity is required (e.g., to handle high loads), the system simulates multiple control schemes (e.g., Scheme A - significantly increase DO; Scheme B - moderately increase MLSS and fine-tune DO). The expected costs C_pred_A and C_pred_B after each scheme are executed are predicted using an activity-cost conversion model, and their cost increments are compared. The scheme with the smallest cost increment is recommended.

[0067] This activity-cost conversion model is not only used for cost monitoring, but also serves as a core engine for forward-looking decision optimization. When faced with process adjustment needs, the system can use the activity-cost conversion model to quickly simulate and compare the cost changes caused by different control paths (such as changing DO, adjusting MLSS, and correcting SRT), thereby automatically identifying and recommending the activity parameter adjustment combination scheme with the lowest total cost, realizing that cost targets directly guide specific process operations.

[0068] like Figure 2 As shown, an embodiment of the present invention provides a system for constructing an activity-cost conversion model, comprising: a first acquisition module, a second acquisition module, and a construction module; The first acquisition module is used to acquire the active feature vector and the overall operating cost; The second acquisition module is used to perform feature engineering construction based on the active feature vector and the comprehensive operating cost to obtain target feature vector pairs; The construction module is used to construct an activity-cost conversion model based on the target feature vector pair and the machine learning algorithm.

[0069] In practical application, the system for constructing the activity-cost conversion model includes a first acquisition module, a second acquisition module, and a construction module. The second acquisition module is connected to both the first and construction modules. After acquiring the activity feature vector and comprehensive operating cost, the first acquisition module transmits these data to the second acquisition module. The second acquisition module performs feature engineering based on the activity feature vector and comprehensive operating cost to obtain target feature vector pairs, which are then transmitted to the construction module. The construction module then constructs the activity-cost conversion model based on the target feature vector pairs and machine learning algorithms. This system constructs the activity-cost conversion model through the cooperation of the first, second, and construction modules. The constructed model establishes a real-time, dynamic mathematical correlation between key technical parameters characterizing sludge health and treatment capacity (such as SOUR, MLSS, and SRT) and the final comprehensive operating cost per ton of water, using a data-driven approach. This ensures that cost control is no longer an independent task parallel to or even conflicting with process optimization, but rather is inherent in every process adjustment decision. The activity-cost conversion model enables the measurement and optimization of technical activities using economic costs as an intuitive benchmark, thereby driving wastewater treatment plants to intelligently and accurately reach a global balance between "stable compliance" and "optimal cost".

[0070] Furthermore, embodiments of this application also disclose an electronic device, Figure 3 This is a structural diagram of an electronic device according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application.

[0071] Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of this application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the activity-cost conversion model construction method disclosed in any of the foregoing embodiments. Furthermore, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0072] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a channel for the construction of the activity-cost conversion model between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to the specific application needs, and is not specifically limited here.

[0073] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon can include operating system 221, computer program 222 and data 223, etc., and the storage method can be temporary storage or permanent storage.

[0074] The operating system 221 manages and controls the various hardware devices on the electronic device 20 and the computer program 222 to enable the processor 21 to perform calculations and processing on the data 223 in the memory 22. It can be Windows Server, Netware, Unix, Linux, etc. The computer program 222, in addition to including a computer program capable of performing the activity-cost conversion model construction method executed by the electronic device 20 as disclosed in any of the foregoing embodiments, may further include computer programs capable of performing other specific tasks. The data 223 may include data received by the activity-cost conversion model construction device from external devices, as well as data collected by its own input / output interface 25.

[0075] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0076] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned method for constructing the activity-cost conversion model. Specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.

[0077] It should be understood that the use of terms such as "method," "apparatus," "unit," and / or "module" in this application is merely to distinguish one method of different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0078] As indicated in this application and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not specifically singular and may include plural forms. Generally, the terms "comprising" and "including" only indicate the inclusion of explicitly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements. An element defined by the phrase "comprising an..." does not exclude the presence of other identical elements in the process, method, product, or apparatus that includes the element.

[0079] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0080] If a flowchart is used in this application, it is used to illustrate the operations performed by the system according to embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0081] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for constructing an activity-cost conversion model, characterized in that, Includes the following steps: Obtain the active feature vector and the overall operating cost; Feature engineering is performed based on the active feature vector and the comprehensive operating cost to obtain target feature vector pairs; An activity-cost conversion model is constructed based on the target feature vector pairs and machine learning algorithms.

2. The method for constructing the activity-cost conversion model according to claim 1, characterized in that, The acquisition of the active feature vector specifically includes: Historical and current operating data are collected and stored in real time through IoT gateways; Extract the time series of characteristic parameters that are directly or strongly correlated with sludge activity from the historical operating data and the current operating data to obtain the activity feature vector.

3. The method for constructing the activity-cost conversion model according to claim 1, characterized in that, The acquisition of comprehensive operating costs specifically includes: The comprehensive operating cost is calculated based on the operating data concurrent with the activity parameters, wherein the specific formula for calculating the comprehensive operating cost is as follows: ; in, This represents the overall operating cost; This represents the electricity consumption cost directly related to biochemical treatment; This indicates the cost of pharmaceutical consumption; This indicates the amount of wastewater treated within the corresponding time period.

4. The method for constructing the activity-cost conversion model according to claim 1, characterized in that, The step of constructing feature engineering based on the active feature vector and the comprehensive operating cost to obtain target feature vector pairs specifically includes: The active feature vector and the comprehensive operating cost are aligned according to a time window to obtain the target active feature vector and the target comprehensive operating cost; Feature derivation and filtering are performed based on the target activity feature vector and the target comprehensive operating cost to obtain target feature vector pairs.

5. The method for constructing the activity-cost conversion model according to claim 1, characterized in that, The step of constructing the activity-cost conversion model based on the target feature vector pair and the machine learning algorithm specifically includes: Build an initial model based on machine learning algorithms; The target feature vector is used to train the initial model to obtain the activity-cost conversion model.

6. The method for constructing the activity-cost conversion model according to claim 1, characterized in that, After constructing the activity-cost conversion model based on the target feature vector pair and the machine learning algorithm, the method further includes: Cost prediction and economic evaluation are performed using the activity-cost conversion model.

7. The method for constructing the activity-cost conversion model according to claim 6, characterized in that, The cost prediction and economic evaluation using the activity-cost conversion model specifically includes: Collect current activity characteristic data and input it into the activity-cost conversion model to obtain the predicted cost per ton of water under the current operating conditions; Calculate the average activity characteristics of water that is stably compliant with standards and within the target time period, and calculate the reference cost corresponding to the average activity characteristics based on the activity-cost conversion model. An economic assessment is conducted based on the predicted cost per ton of water and the reference cost.

8. A system for constructing an activity-cost conversion model, characterized in that, include: The module consists of a first acquisition module, a second acquisition module, and a construction module. The first acquisition module is used to acquire the active feature vector and the overall operating cost; The second acquisition module is used to perform feature engineering construction based on the active feature vector and the comprehensive operating cost to obtain target feature vector pairs; The construction module is used to construct an activity-cost conversion model based on the target feature vector pair and the machine learning algorithm.

9. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor, the memory storing a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The storage medium is used to store a computer program that causes a computer to perform the method described in any one of claims 1-7.

Images