Water quality purification plant operation management two-dimensional evaluation method and system

Abstract

The application discloses a water quality purification plant operation management two-dimension evaluation method and system, relates to the technical field of operation management, and comprises the following steps: collecting a one-dimensional array of a target water quality purification plant in a preset period to obtain a periodic index weight; taking the marginal value coefficient and incentive weight of each operation index in the region where the plant is located as a benchmark to identify a value-performance gap; constructing an efficiency frontier envelope surface to determine an operation evaluation result; obtaining a combined improvement strategy through decision-making; and performing operation management alarm and optimization management guidance. The application solves the technical problems that the prior art relies on only single-plant self data to carry out efficiency evaluation, lacks regional dimension value constraints, and has one-sided evaluation results and poor adaptability of optimization suggestions, and achieves two-dimension fusion evaluation of plant internal data and regional examination value and resource iterative allocation, thereby improving the technical effects of water quality purification plant operation efficiency evaluation accuracy and optimization guidance targeting.

Description

Technical Field

[0001] This invention relates to the field of operation and management technology, specifically to a two-dimensional evaluation method and system for the operation and management of water purification plants. Background Technology

[0002] Currently, most evaluations of the operational efficiency of water purification plants are based solely on the analysis and judgment of the plant's own operational data. They lack the constraints of multiple plant samples in the region and the regional assessment value dimension. This makes it difficult to take into account the matching relationship between the actual operating status of the plant and the regional environmental incentive weights, which can easily lead to one-sided and distorted evaluation results. At the same time, traditional evaluation methods only output efficiency conclusions and cannot complete the resource coordination and iterative optimization between input and output. The optimization guidance suggestions output are not adaptable enough and cannot provide operation and maintenance personnel with a basis for refined operation and management that can be implemented.

[0003] Existing technologies rely solely on data from a single plant for efficiency evaluation, lacking regional value constraints, resulting in biased evaluation results and poor adaptability of optimization suggestions. Summary of the Invention

[0004] This application provides a two-dimensional evaluation method and system for the operation and management of water purification plants, which addresses the technical problems of existing technologies that rely solely on data from a single plant for efficiency evaluation, lack regional dimension value constraints, and result in biased evaluation results and poor adaptability of optimization suggestions.

[0005] In view of the above problems, this application provides a two-dimensional evaluation method and system for the operation and management of water purification plants.

[0006] The first aspect of this application provides a two-dimensional evaluation method for the operation and management of water purification plants, the method comprising: A one-dimensional data matrix of the target water purification plant is collected at a preset period. The weights of periodic indicators are obtained through normalized index entropy analysis. A two-dimensional data matrix of the target water purification plant is collected at the same preset period. The value-performance gap is identified based on the marginal value coefficients and incentive weights of various operational indicators in the plant's location. Input-output combinations are defined using input indicators, expected output indicators, and unexpected output indicators. An efficiency frontier envelope is constructed based on the one-dimensional data matrix, periodic indicator weights, the two-dimensional data matrix, and the value-performance gap. Distance is measured between this envelope and the expected standard efficiency frontier envelope to determine the operational evaluation results. Based on these operational evaluation results, a combination improvement strategy is determined, with resource transfer coordination between input-output combinations as the first-order optimization objective and the improvement of the correlation between each input-output combination as the second-order optimization objective. The operational evaluation results and the combination improvement strategy are integrated to provide operational management alerts and optimization management guidance.

[0007] A second aspect of this application provides a two-dimensional evaluation system for the operation and management of a water purification plant, the system comprising: The system comprises the following modules: a periodic indicator weight acquisition module, which collects a one-dimensional data matrix of the target water purification plant at a preset period and obtains the periodic indicator weights through normalized indicator entropy analysis; a gap identification module, which collects a two-dimensional data matrix of the target water purification plant at a preset period and identifies the value-performance gap based on the marginal value coefficients and incentive weights of various operational indicators in the plant's location; an operational evaluation result determination module, which defines input-output combinations based on input indicators, expected output indicators, and unexpected output indicators, constructs an efficiency frontier envelope based on the one-dimensional data matrix, periodic indicator weights, two-dimensional data matrix, and value-performance gap, and determines the operational evaluation result by measuring the distance between the envelope and the expected standard efficiency frontier envelope; a combined improvement strategy acquisition module, which, based on the operational evaluation results, determines a combined improvement strategy by using resource transfer coordination between input-output combinations as the first-order optimization objective and improving the correlation of each input-output combination as the second-order optimization objective; and an operational management alarm module, which integrates the operational evaluation results and the combined improvement strategy to provide operational management alarms and optimization management guidance.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: A one-dimensional data matrix of the target water purification plant is collected at a preset period to obtain the weights of periodic indicators. Using the marginal value coefficients and incentive weights of various operational indicators in the plant's location as a benchmark, the value-performance gap is identified. An efficiency frontier envelope is constructed, and its distance from the expected standard efficiency frontier envelope is measured to determine the operational evaluation results. Using resource transfer coordination between input-output combinations as the first-order optimization objective and improving the correlation between each input-output combination as the second-order optimization objective, a combined improvement strategy is determined. The operational evaluation results and combined improvement strategies are integrated to provide operational management alerts and optimization management guidance. This achieves a two-dimensional fusion evaluation of water purification plant data and regional assessment value, as well as iterative resource allocation, improving the accuracy of water purification plant operational efficiency evaluation and the targeted nature of optimization guidance. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 A schematic diagram of a two-dimensional evaluation method for the operation and management of a water purification plant, provided in an embodiment of this application; Figure 2 This is a schematic diagram of a two-dimensional evaluation system for the operation and management of a water purification plant, provided as an embodiment of this application.

[0011] Figure labeling: Periodic indicator weight acquisition module 10, gap identification module 20, operation evaluation result determination module 30, combined improvement strategy acquisition module 40, operation management alarm module 50. Detailed Implementation

[0012] This application provides a two-dimensional evaluation method and system for the operation and management of water purification plants, which addresses the technical problems of existing technologies that rely solely on data from a single plant for efficiency evaluation, lack regional dimension value constraints, and result in biased evaluation results and poor adaptability of optimization suggestions.

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0014] Example 1, as Figure 1 As shown, this application provides a two-dimensional evaluation method for the operation and management of water purification plants, the method comprising: Step S100: Collect a one-dimensional array of data from the target water purification plant at a preset period, and obtain the periodic index weights through normalized index entropy value analysis.

[0015] Specifically, periodic indicator collection is conducted on the target water purification plant according to a pre-set evaluation cycle. This includes periodic operational indicators such as pollutant removal rate, electricity consumption per ton of water, and chemical consumption per ton of water. The collected data are then organized to construct a one-dimensional matrix. Normalization preprocessing is performed on the raw data of each indicator within the one-dimensional matrix to eliminate computational interference caused by differences in dimensions and numerical magnitudes between different indicators. Based on the normalized indicator dataset, the entropy value corresponding to each operational indicator is calculated. According to the rule that a smaller entropy value indicates a higher objective weight, the objective weight of each indicator is determined. The pre-stored standard weight distribution is retrieved, and the calculated objective weights are used as the basis for entropy value adjustment of the standard weight distribution. This completes the fusion and calibration of objective weights and industry standard weights, ultimately outputting periodic indicator weights adapted to the current operating conditions, providing a basis for calculating indicator weights for the subsequent construction of the efficiency frontier envelope.

[0016] Step S200: Collect a two-dimensional data matrix of the target water purification plant at a preset period, and identify the value-performance gap based on the marginal value coefficient and incentive weight of each operating indicator in the region where the plant is located.

[0017] Specifically, the administrative region where the target water purification plant is located is determined, and the marginal value coefficients and incentive weights corresponding to operational indicators such as COD reduction rate, ammonia nitrogen reduction rate, and influent COD concentration adjustment coefficient within this region are obtained. Following a pre-set evaluation period consistent with [previous standards], sample data of the corresponding operational indicators for each water purification plant within the region are collected and organized into a two-dimensional matrix. Based on this two-dimensional matrix, marginal value curves and weight curves corresponding to each operational indicator are generated. The marginal value curve represents the relationship between indicator returns and indicator performance, while the weight curve represents the allocation of incentive weights to each operational indicator at the regional level. A comparative analysis of the two curves is conducted, and items with high incentive weights but low actual indicator scores are selected as gap items. This identifies the value-performance gap, introducing regional incentive-oriented evaluation constraints for the subsequent construction of the efficiency frontier envelope.

[0018] Step S300: Define the input-output combination using input indicators, expected output indicators, and unexpected output indicators. Construct the efficiency frontier envelope based on a one-dimensional matrix, periodic indicator weights, a two-dimensional matrix, and the value-performance gap. Measure the distance between the efficiency frontier envelope and the expected standard efficiency frontier envelope to determine the operational evaluation results.

[0019] Specifically, input indicators, expected output indicators, and unexpected output indicators are used as benchmarks to define input-output combinations and set them as vertices of the envelope surface. A first efficiency front envelope surface, i.e., the expected standard efficiency front envelope surface, is constructed where the vertex efficiency value is always 1. Then, a one-dimensional matrix, periodic indicator weights, a two-dimensional matrix, and the value-performance gap are introduced as constraints. The objective function for efficiency is to maximize the weighted ratio of output to input. A linear programming solution is performed with the constraint that the weighted sum of outputs for all input-output combinations does not exceed the weighted sum of inputs. The vertex efficiency value corresponding to each input-output combination is calculated, and the vertex efficiency is then determined. Vertices with an efficiency value of 1 are designated as leading-edge vertices, while vertices with an efficiency value less than 1 are placed inside the leading-edge surface. Adjacent leading-edge vertices are connected via a linear envelope to form a second efficiency leading-edge envelope. Then, a mapping distance calculation is performed on the first and second efficiency leading-edge envelopes to obtain the distance vector from each vertex inside the leading-edge surface to the leading-edge surface. This distance vector is obtained by jointly measuring the input redundancy, the expected output deficiency, and the unexpected output excess. Based on the obtained distance vectors, the operational evaluation results of the target water purification plant are comprehensively calculated and output, providing a quantitative evaluation basis for the subsequent generation of combined improvement strategies.

[0020] Step S400: Based on the operational evaluation results, with the coordination of resource transfer between input-output combinations as the first-order optimization objective and the improvement of the correlation between each input-output combination as the second-order optimization objective, a combination improvement strategy is determined.

[0021] Specifically, the system reads the input redundancy, expected output shortfall, and unexpected output excess corresponding to each input-output combination from the operational evaluation results. It prioritizes resource transfer coordination between input-output combinations as the first-order optimization objective. Based on the redundancy and deficit of each combination, it performs cross-combination resource allocation calculations to obtain the first resource coordination strategy. Then, it takes the correlation improvement of each input-output combination as the second-order optimization objective, retrieves the set of improvement strategies associated with each input-output combination, and makes control decisions based on the first resource coordination strategy and the correlation constraints between indicators. It performs correlation adaptation correction on the resource allocation scheme, eliminates conflicts between improvement actions of different input-output combinations, and finally outputs a combination improvement strategy adapted to the actual operating conditions of the target water purification plant. Subsequently, iterative verification can be carried out by updating the second efficiency frontier envelope based on this combination improvement strategy.

[0022] Step S500: Integrate the operational evaluation results with combined improvement strategies to provide operational management alerts and optimization management guidance.

[0023] Specifically, the system integrates the quantitative operational evaluation results and the combined improvement strategies of the decision outputs, pre-sets multi-level alarm thresholds corresponding to the operational evaluation results, and compares the operational evaluation results of the target water purification plant with the multi-level alarm thresholds item by item. When the operational evaluation result falls into the corresponding alarm threshold range, the corresponding level of operational management alarm is triggered. Simultaneously, the combined improvement strategies are broken down into executable optimization items for input indicators, expected output indicators, and unexpected output indicators, forming targeted optimization management guidance information. The alarm level, evaluation score, and each executable optimization item are correlated and output, providing water purification plant operation and maintenance management personnel with intuitive feedback on the current operational level and specific operation and maintenance adjustment directions, facilitating operation and maintenance personnel to carry out plant operation and control adjustments based on the output content.

[0024] In one possible implementation, step S100 further includes: Step S110: According to the preset cycle, periodic indicators are collected from the target water purification plant to obtain a one-dimensional data matrix. The periodic indicators include pollutant removal rate, electricity consumption per ton of water, and chemical consumption per ton of water.

[0025] Step S120: After normalizing the one-dimensional matrix, calculate the index entropy value and determine the objective weight of the index. The smaller the entropy value, the higher the objective weight.

[0026] Step S130: Call the standard weight distribution, and perform entropy value adjustment on the standard weight distribution according to the objective weight of the indicator to obtain the periodic indicator weight.

[0027] Specifically, according to a pre-set evaluation cycle, the actual operation data of the water purification plant is periodically read from the plant's production and operation database. Sample data of three periodic indicators—pollutant removal rate, electricity consumption per ton of water, and chemical consumption per ton of water—are collected within the cycle. The pollutant removal rate represents the plant's ability to remove pollutants from wastewater, electricity consumption per ton of water is the electricity consumed to treat each ton of wastewater, and chemical consumption per ton of water is the amount of chemicals consumed to treat each ton of wastewater. The measured data of the above three indicators within the same evaluation cycle are arranged in order according to the indicator category to construct a one-dimensional matrix composed of the sample data of each indicator within the cycle. This one-dimensional matrix records the original sequence of the basic operating indicators of the target water purification plant in the current cycle, providing raw data input for subsequent normalization treatment and indicator entropy analysis.

[0028] The output one-dimensional matrix is ​​read, and for positive indicators such as pollutant removal rate and negative indicators such as water-electricity consumption per ton and chemical consumption per ton, the corresponding normalization calculation function is selected to perform dimensionless calculation. The original values ​​of all indicators in the one-dimensional matrix are mapped to the 0-1 value range to obtain a normalized matrix. This avoids the interference of different measurement units and numerical magnitudes of different indicators on subsequent entropy value calculation. Based on the distribution probability of sample data under each indicator in the normalized matrix, the entropy value of each periodic indicator is calculated in turn. The entropy value can reflect the dispersion of indicator data. The greater the dispersion of indicator data, the smaller the corresponding entropy value, the greater the evaluation information carried by the indicator, and the higher its objective weight. According to the rule that the smaller the entropy value, the higher the objective weight, the objective weight of each indicator corresponding to pollutant removal rate, water-electricity consumption per ton, and chemical consumption per ton is obtained by combining the entropy value calculation results. This provides objective weight calculation results for subsequent entropy value weighting calculation.

[0029] The pre-stored industry standard weight distribution for water purification plants is retrieved. This standard weight distribution is based on industry standards and operational experience, with preset baseline weight allocation results for pollutant removal rate, electricity consumption per ton of water, and chemical consumption per ton of water. The calculated objective weights of each indicator are used as the input for weight adjustment. An entropy fusion weight adjustment algorithm is used to integrate the discrete characteristics of the actual data of the plant reflected by the objective weights into the standard weight distribution. The baseline weights of each indicator within the standard weight distribution are scaled and corrected to complete the fusion and calibration of the industry prior experience weights and the data-driven objective weights. Then, the weights of each indicator after weight adjustment are normalized to ensure that the sum of all indicator weights is 1. Finally, the periodic indicator weights that are adapted to the actual operating conditions of the plant in the current evaluation period are output, providing a weighted evaluation basis for the subsequent construction of the efficiency frontier envelope.

[0030] In one possible implementation, step S200 further includes: Step S210: For the region where the plant is located, determine the marginal value coefficient and incentive weight of each operational indicator, including COD reduction rate, ammonia nitrogen reduction rate and influent COD concentration adjustment coefficient.

[0031] Step S220: Based on the preset period, collect the binary data matrix of each water purification plant for the operational indicators, and determine the marginal value curve and weight curve.

[0032] Step S230: Identify the value-performance gap based on the marginal value curve and the weight curve, wherein the value-performance gap is the gap item with high weight and low score in the curve.

[0033] Specifically, this involves acquiring environmental assessment policies, regional pollutant reduction records, and historical operational sample data from multiple water purification plants in the region where the target water purification plant is located. For the marginal value coefficient, the regional pollutant reduction benefit corresponding to a unit percentage increase in the indicator is used as the accounting basis. The ratio of the change in regional environmental benefits to the change in the corresponding operational indicator is calculated, yielding the marginal value coefficients for COD reduction rate, ammonia nitrogen reduction rate, and influent COD concentration adjustment coefficient. For the incentive weight, the assessment scores corresponding to each operational indicator in the local environmental assessment documents are extracted, and the original assessment scores of each indicator are normalized by dividing the original assessment scores of all indicators by the sum of all indicator assessment scores. The conversion yields the incentive weights corresponding to each operational indicator. For example, if a local plant assigns a very high weight to the COD reduction rate and a low weight to the sludge treatment rate, but the target plant's operational data shows that sludge treatment costs continue to exceed budgets while COD reduction remains stable and meets targets, it is determined that the plant is over-investing in low-value areas. When generating optimization suggestions, some of the investment resources can be transferred from sludge treatment management to further improving the stability of the COD reduction rate. The marginal value coefficients and incentive weights of each operational indicator obtained through the above calculations are indexed and stored to provide regional-level benchmark calculation parameters for subsequent fitting of marginal value curves and weight curves based on two-dimensional matrix to identify the value-performance gap.

[0034] Using the same preset evaluation period as in step S100, the measured data of COD reduction rate, ammonia nitrogen reduction rate, and influent COD concentration adjustment coefficient of all water purification plants within the target plant area are collected. This data is then organized and constructed into a two-dimensional matrix containing multiple plant areas and multiple operational indicators. The actual performance of each plant's operational indicators in the two-dimensional matrix is ​​used as the abscissa, and the corresponding environmental marginal benefit value is calculated using the already solved marginal value coefficient as the gain coefficient as the ordinate. A marginal value curve is generated through discrete sample point interpolation fitting. This curve characterizes the correspondence between changes in operational indicator performance and regional environmental marginal benefits. Then, using the operational indicator performance of each plant area in the two-dimensional matrix as the abscissa and the normalized incentive weight as the ordinate, each sample point is mapped and interpolated to obtain a weight curve. This curve reflects the distribution relationship of regional assessment incentive weights with indicator performance. The fitted marginal value curve and weight curve are output and saved, providing a curve analysis basis for subsequent comparison and analysis to identify the value-performance gap.

[0035] Read the fitted marginal value curve and weight curve, and map the actual scores of each operational indicator of the target water purification plant in the current cycle onto the two curves respectively. Obtain the actual marginal benefit value and incentive weight value at the corresponding indicator position. Pre-set the weight threshold and indicator score threshold. Incentive weight greater than the weight threshold is judged as high weight item, and actual score of indicator less than the score threshold is judged as low score item. Iterate through each operational indicator such as COD reduction rate, ammonia nitrogen reduction rate, and influent COD concentration adjustment coefficient, and screen the evaluation items that meet both the high weight and low score conditions as gap items, i.e., value-performance gap. This gap indicates that the operational indicator has high assessment incentive value at the regional level, but the actual operation performance of the plant does not match this value positioning. Mark and record all identified value-performance gaps as constraints input into the subsequent efficiency frontier envelope construction process.

[0036] In one possible implementation, step S300 further includes: Step S310: For the target water purification plant, based on the input indicators, expected output indicators and unexpected output indicators, define the input-output combination as the vertex and construct the first efficiency front envelope, wherein the vertex efficiency value of the first efficiency front envelope is always 1.

[0037] Step S320: Construct a second efficiency frontier envelope based on the one-dimensional matrix, periodic index weights, two-dimensional matrix, and value-performance gap.

[0038] Step S330: Determine the operation evaluation result based on the first efficiency front envelope and the second efficiency front envelope.

[0039] Specifically, the consumption of reagents, electricity consumption, and manpower operation and maintenance costs are selected as input indicators; the reduction of COD and ammonia nitrogen are selected as expected output indicators; and the amount of wastewater exceeding standards and the amount of sludge generated are selected as undesired output indicators. The multi-dimensional data of the above inputs, expected outputs, and undesired outputs are paired and combined to form multiple sets of input-output combinations. Each set of input-output combinations is used as a vertex in the geometric space of data envelopment analysis. The first efficiency front envelope is the standard efficiency front envelope under the theoretical ideal state, representing the ideal boundary where input and output reach the optimal match. The efficiency value of all vertices falling on the front envelope is set to be always 1. All ideal optimal vertices are connected by the linear envelope algorithm of data envelopment analysis to form a first efficiency front envelope with a closed convex hull structure, which is used as a theoretical reference benchmark for the contrast of the second efficiency front envelope under subsequent actual working conditions.

[0040] The model uses a one-dimensional matrix, periodic index weights, a two-dimensional matrix, and the identified value-performance gap as input constraints. The objective function is to maximize the weighted ratio of output to input. A linear programming solution is implemented, with the constraint that the weighted sum of outputs for all input-output combinations does not exceed the weighted sum of inputs. The vertex efficiency value for each input-output combination is calculated, and combinations with an efficiency value of 1 are designated as frontier vertices. Combinations with efficiency values ​​less than 1 are placed within the frontier surface. A linear envelope method is then used to sequentially connect adjacent frontier vertices, forming a convex hull structure. This constructs a second efficiency frontier envelope that fits the actual operation of the target water purification plant while considering regional value gap constraints. This second envelope is used for subsequent distance measurement calculations with the first efficiency frontier envelope.

[0041] The first efficiency frontier envelope, representing the theoretical ideal state, and the second efficiency frontier envelope, constructed based on the actual operating conditions of the plant, are obtained. Spatial mapping distance calculations are performed on the two efficiency frontier envelopes to obtain the distance vectors from the vertices of each input-output combination within the second efficiency frontier envelope to the first efficiency frontier envelope. These distance vectors are measured by a combination of input redundancy, expected output shortfall, and unexpected output overshoot. The larger the magnitude of the distance vector, the more severe the deviation of the input-output combination from the theoretical optimal operating level. Based on obtaining the input redundancy, expected output shortfall, and unexpected output overshoot corresponding to each vertex, a weighted fusion calculation is performed by combining the distance vector magnitude and the periodic index weights to quantitatively output the operational evaluation results of the target water purification plant. These operational evaluation results can comprehensively reflect the overall operational efficiency of the plant and the degree of deviation of various indicators, providing a quantitative calculation basis for subsequent generation of input-output combination improvement strategies.

[0042] In one possible implementation, step S320 further includes: Step S321: Determine the vertex efficiency value based on the one-dimensional matrix, periodic index weights, two-dimensional matrix, and value-performance gap.

[0043] Step S322: Vertices with a vertex efficiency value of 1 are located on the leading edge surface, vertices with a vertex efficiency value less than 1 are located inside the leading edge surface, and vertices located on the leading edge surface are taken as leading edge vertices. Adjacent leading edge vertices are connected by a linear envelope to form the second efficiency leading edge envelope surface.

[0044] Specifically, the measured periodic indicators and periodic indicator weights of the one-dimensional array of the factory area will be used. Two-dimensional matrix of regional samples, gap penalty coefficient corresponding to the value-performance gap. Input a linear programming model with the objective function of maximizing the weighted ratio of output to input: In the formula, e is the vertex efficiency value to be solved. Let i be the input indicator. Let r be the output indicator. , Input and output weights are respectively used, and periodic indicator weights are introduced. right , Initialize the weights and simultaneously introduce a gap penalty coefficient for the indicator corresponding to the value-performance gap. Revise the output terms to increase the constraint effect of high-weight, low-score gap indicators on efficiency calculations; set constraints: , where m represents the entire input-output combination sample, ensuring that the weighted sum of outputs for all samples does not exceed the weighted sum of inputs. , Ensure the weights are non-negative; for each input-output combination vertex, substitute the above objective function and all constraints to perform a linear programming solution. The optimal solution for the corresponding vertex is the vertex efficiency value e for that input-output combination. This means that the sample has reached a relatively optimal state under the current constraints. This indicates an operational efficiency loss. The calculated efficiency values ​​of all vertices provide a quantitative basis for distinguishing between frontier vertices and internal vertices and constructing a second efficiency frontier envelope.

[0045] The vertex efficiency values ​​corresponding to each input-output combination obtained from the solution are read, and all spatial vertices are classified and judged. Vertices of input-output combinations with a vertex efficiency value equal to 1 are judged as frontier vertices and placed at the boundary of the frontier surface. Vertices of input-output combinations with a vertex efficiency value less than 1 are judged as non-frontier vertices and assigned to the internal region of the frontier surface. Based on the geometric adjacency relationship of each vertex in the input-output multidimensional space, mutually adjacent frontier vertices are selected. Using the convex hull linear envelope algorithm of data envelopment analysis, linear piecewise hyperplanes are constructed for each pair of adjacent frontier vertices. The closed convex hull structure is formed by splicing all the piecewise hyperplanes. This constitutes the second efficiency frontier envelope surface that takes into account the measured data of the plant area, regional samples, and value-performance gap constraints. This envelope surface represents the operational efficiency boundary that can be achieved under actual working conditions and regional evaluation constraints, and provides a geometric model for subsequent distance vector measurement with the first efficiency frontier envelope surface.

[0046] In one possible implementation, step S321 further includes: For each defined input-output combination, the objective function is to maximize the weighted ratio of output to input as the efficiency value. The constraint is that the weighted sum of output of all input-output combinations does not exceed the weighted sum of input. Linear programming is used to solve the problem based on a one-dimensional matrix, periodic index weights, a two-dimensional matrix, and the value-performance gap to obtain the vertex efficiency value of each input-output combination.

[0047] Specifically, the measured input-output data of the target plant in the one-dimensional matrix and the weights of periodic indicators are used as the initial basis for assigning input-output weights. Input-output samples from other water purification plants in the region within the two-dimensional matrix are introduced to expand the comparison set. Simultaneously, penalty correction terms are added to high-weight, low-score indicators corresponding to the value-performance gap, thus constructing the objective function. Where e is the vertex efficiency value. As input indicators, As an output indicator, Set constraints for input and output weights respectively. m iterates through all input-output combination samples, ensuring that the weighted sum of outputs for all combinations is no greater than the weighted sum of inputs, and that the weights are non-negative. , For each input-output combination, the above objective function and constraints are substituted sequentially to perform linear programming operations, complete the optimization solution of weights, and output the vertex efficiency value corresponding to each group, thereby quantifying the relative operating efficiency of each input-output combination under multi-source data and gap constraints.

[0048] In one possible implementation, step S330 further includes: Step S331: Perform mapping distance calculation on the first efficiency front envelope and the second efficiency front envelope to obtain the distance vector from the internal vertex of each front to the front. The distance vector is based on a joint metric of input redundancy, expected output insufficiency, and unexpected output overscalar.

[0049] Step S332: Solve for the operation evaluation results based on each distance vector.

[0050] Specifically, a multi-dimensional spatial projection operation is performed on each input-output combination vertex inside the second efficiency front envelope surface to the theoretical benchmark first efficiency front envelope surface. The optimal projection point corresponding to each vertex on the first efficiency front envelope surface is solved by radial projection optimization. By comparing the coordinates of the internal vertex and the coordinates of the projection point, the input redundancy in the input dimension, the expected output deficiency in the expected output dimension, and the unexpected output excess in the unexpected output dimension are solved respectively. The input redundancy is the difference between the input component of the internal vertex and the corresponding input component of the projection point. The expected output deficiency is the difference between the expected output component of the projection point and the expected output component of the internal vertex. The unexpected output excess is the difference between the unexpected output component of the internal vertex and the unexpected output component of the projection point. The above three types of differences are used as independent components of the multi-dimensional distance vector and combined to form a complete distance vector. Each internal vertex corresponds to a set of independent distance vectors. The magnitude of each component of the distance vector represents the degree to which the corresponding indicator dimension deviates from the theoretical optimal boundary. All the distance vector data obtained are output to the subsequent stage for the quantitative solution of the operation evaluation results.

[0051] The multi-dimensional distance vectors corresponding to each internal vertex are read from the calculation. The distance vectors contain three types of dimensional components: input redundancy, expected output shortfall, and unexpected output overshoot. The obtained periodic index weights are retrieved to weight each dimensional component. Positive penalty calculation is performed on input redundancy, and gap quantification calculation is performed on expected output shortfall and unexpected output overshoot. The vector magnitude of each weighted component is obtained. This magnitude represents the overall deviation of the actual operating conditions of the target water purification plant from the theoretical optimal first efficiency frontier envelope. The mapping conversion relationship between the magnitude and the evaluation score is pre-configured, and the weighted vector magnitude is converted into a normalized operation evaluation value. At the same time, the deviation values ​​of each dimension are retained. The overall evaluation value and the deviation values ​​of each component together constitute the operation evaluation result, which can reflect the overall operation efficiency level of the plant and locate the specific deviations of each input and output indicator. This provides a complete quantitative basis for determining the first-order and second-order optimization objectives and generating combined improvement strategies.

[0052] In one possible implementation, step S400 further includes: Step S410: Based on the input redundancy, expected output shortfall, and unexpected output excess corresponding to each input-output combination, resource coordination is performed with the resource transfer target between combinations to obtain the first resource coordination strategy.

[0053] Step S420: Based on the first resource coordination strategy, and based on the set of improvement strategies associated with each input-output combination, make control decisions to obtain a combined improvement strategy.

[0054] Specifically, the system batch-reads the input redundancy, expected output deficiency, and unexpected output excess of each input-output combination calculated using the frontier distance vector. It then precisely identifies surplus input-output combinations with redundant inputs and available idle resources, as well as deficit input-output combinations with output gaps, exceeding targets, and requiring resource replenishment. With optimal cross-combination resource transfer matching as the core objective, it establishes a resource supply-demand matching mapping relationship. It prioritizes the targeted allocation of redundant input resources such as electricity, chemicals, and maintenance personnel from surplus combinations to deficit combinations, specifically addressing the expected output deficiency and suppressing unexpected output excess in deficit combinations. Simultaneously, it adheres to the constraint of constant total resources and dynamic cross-combination redistribution, eliminating the contradiction between resource idleness and resource shortage in each independent input-output combination. This completes the overall transfer and balanced adaptation of all available resources, forming a first-order resource coordination strategy centered on inter-combination resource transfer coordination, which serves as the basic optimization scheme corresponding to the first-order optimization objective.

[0055] The system retrieves the first resource coordination strategy, which determines the direction and scale of resource transfer between different input-output combinations. Simultaneously, it reads a pre-constructed set of improvement strategies associated with each input-output combination. This set stores multiple executable control measures for various deviation scenarios, including input redundancy, insufficient expected output, and excessive undesired output. Prioritizing these control measures based on the value-performance gap constraint, it prioritizes those that improve high-weight, low-score gap indicators. The resource transfer constraints of the first resource coordination strategy are coupled and matched with the control measures within the strategy set. For each input-output combination, the system determines the resource allocation ratio, indicator adjustment direction, and control execution magnitude, eliminating conflicting control actions. The system integrates and merges these control measures to form a combined improvement strategy that considers both cross-combination resource reallocation and single-group indicator deviation correction. This strategy clarifies the specific adjustment plans for each input, expected output, and undesired output indicator, providing complete decision-making output for subsequent operational management alerts and optimization guidance.

[0056] In one possible implementation, step S420 further includes: Step S421: Based on the combined improvement strategy, update the second efficiency front envelope and verify whether it meets the standard. If there are vertices inside the front, perform iterative optimization until convergence and update the combined improvement strategy.

[0057] Specifically, based on the generated combined improvement strategy, the input redundancy, expected output shortfall, and unexpected output excess of each input-output combination of the target water purification plant are synchronously corrected and updated. The corrected new input-output sample data are then resubmitted into the vertex efficiency value solution model to recalculate the vertex efficiency value corresponding to each input-output combination. Furthermore, according to the efficiency value judgment rules, the leading vertices are re-selected, and the second efficiency leading edge envelope is fitted and updated. After the iterative update of the leading edge envelope is completed, convergence verification is performed, ensuring that there are no internal vertices with an efficiency value less than 1 within the updated second efficiency leading edge envelope. As a convergence criterion, if there are still inefficient vertices within the envelope that do not meet the standard, the optimization is deemed unsuccessful. The system automatically returns to the resource coordination and strategy matching process to readjust the resource allocation ratio and optimize the priority of control measures. Iteratively, a new combined improvement strategy is generated and the second efficiency front envelope is updated again. This process continues to iterate and correct operational deviations one by one until the efficiency value of all vertices in the second efficiency front envelope is 1, there are no inefficient vertices within the envelope, and the model reaches a stable convergence state. Finally, the optimal combined improvement strategy after iterative convergence is output, realizing closed-loop optimization and control of the water purification plant's operational efficiency.

[0058] Example 2, based on the same inventive concept as the two-dimensional evaluation method for the operation and management of a water purification plant in the foregoing examples, such as... Figure 2 As shown, this application provides a two-dimensional evaluation system for the operation and management of water purification plants. The system and method embodiments in this application are based on the same inventive concept. The system includes: The periodic indicator weight acquisition module 10 is used to collect a one-dimensional array of data from the target water purification plant at a preset period, and obtain the periodic indicator weights through normalized indicator entropy value analysis.

[0059] The gap identification module 20 is used to collect a two-dimensional array of data of the target water purification plant at a preset period, and to identify the value-performance gap based on the marginal value coefficient and incentive weight of each operating indicator in the region where the plant is located.

[0060] The Operation Evaluation Result Determination Module 30 is used to define input-output combinations based on input indicators, expected output indicators, and unexpected output indicators. It constructs an efficiency frontier envelope based on a one-dimensional matrix, periodic indicator weights, a two-dimensional matrix, and the value-performance gap. It measures the distance between the efficiency frontier envelope and the expected standard efficiency frontier envelope to determine the operation evaluation result.

[0061] The combined improvement strategy acquisition module 40 is used to determine the combined improvement strategy based on the operational evaluation results, with the resource transfer coordination between input-output combinations as the first-order optimization objective and the correlation improvement of each input-output combination as the second-order optimization objective.

[0062] The operations management alarm module 50 is used to integrate the operations evaluation results and combined improvement strategies to provide operations management alarms and optimization management guidance.

[0063] Furthermore, the system is also used to implement the following functions: According to a preset period, periodic indicators are collected from the target water purification plant to obtain a one-dimensional data matrix. The periodic indicators include pollutant removal rate, electricity consumption per ton of water, and chemical consumption per ton of water. After normalizing the one-dimensional data matrix, the entropy value of the indicators is calculated to determine the objective weight of the indicators. The smaller the entropy value, the higher the objective weight. The standard weight distribution is called, and the entropy value is adjusted according to the objective weight of the indicators to obtain the weight of the periodic indicators.

[0064] Furthermore, the system is also used to implement the following functions: For the region where the plant is located, the marginal value coefficient and incentive weight of each operational indicator are determined. The operational indicators include COD reduction rate, ammonia nitrogen reduction rate, and influent COD concentration adjustment coefficient. According to a preset period, for the operational indicators, a binary data matrix of each water purification plant is collected to determine the marginal value curve and weight curve. Based on the marginal value curve and weight curve, the value-performance gap is identified, where the value-performance gap is the gap item with high weight and low score in the curve.

[0065] Furthermore, the system is also used to implement the following functions: For the target water purification plant, based on input indicators, expected output indicators, and unexpected output indicators, an input-output combination is defined as the vertex to construct a first efficiency frontier envelope, wherein the vertex efficiency value of the first efficiency frontier envelope is always 1; based on the one-dimensional matrix, periodic indicator weights, two-dimensional matrix, and value-performance gap, a second efficiency frontier envelope is constructed; based on the first efficiency frontier envelope and the second efficiency frontier envelope, the operational evaluation result is determined.

[0066] Furthermore, the system is also used to implement the following functions: Based on the one-dimensional matrix, periodic index weights, two-dimensional matrix, and value-performance gap, the vertex efficiency value is determined; vertices with an efficiency value of 1 are located on the leading edge, vertices with an efficiency value less than 1 are located inside the leading edge, and vertices located on the leading edge are designated as leading edge vertices. Adjacent leading edge vertices are connected by a linear envelope to form the second efficiency leading edge envelope.

[0067] Furthermore, the system is also used to implement the following functions: For each defined input-output combination, the objective function is to maximize the weighted ratio of output to input as the efficiency value. The constraint is that the weighted sum of output of all input-output combinations does not exceed the weighted sum of input. Linear programming is used to solve the problem based on a one-dimensional matrix, periodic index weights, a two-dimensional matrix, and the value-performance gap to obtain the vertex efficiency value of each input-output combination.

[0068] Furthermore, the system is also used to implement the following functions: The mapping distance between the first efficiency front envelope and the second efficiency front envelope is calculated to obtain the distance vector from the internal vertex of each front to the front. The distance vector is based on a joint measure of input redundancy, expected output deficiency, and unexpected output excess. The operation evaluation result is solved based on each distance vector.

[0069] Furthermore, the system is also used to implement the following functions: Based on the input redundancy, expected output shortfall, and unexpected output excess corresponding to each input-output combination, resource coordination is carried out with the resource transfer target between combinations to obtain the first resource coordination strategy; based on the first resource coordination strategy and the set of improvement strategies associated with each input-output combination, control decisions are made to obtain the combination improvement strategy.

[0070] Furthermore, the system is also used to implement the following functions: Based on the combined improvement strategy, the second efficiency front envelope is updated, and it is verified whether the target is met. If there are vertices inside the front, iterative optimization is performed until convergence, and the combined improvement strategy is updated.

[0071] It should be noted that the order of the embodiments described above is for descriptive purposes only and does not represent the superiority or inferiority of the embodiments. Specific embodiments of this specification have been described above. Furthermore, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0072] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0073] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application intends to include such modifications and variations.

Claims

1. A two-dimensional evaluation method for the operation and management of water purification plants, characterized in that, The method includes: A one-dimensional array of data from the target water purification plant is collected at a preset period. The weights of the periodic indicators are obtained by analyzing the normalized index entropy values. A two-dimensional data matrix of the target water purification plant is collected at a preset period. The value-performance gap is identified based on the marginal value coefficient and incentive weight of each operating indicator in the region where the plant is located. Input-output combinations are defined using input indicators, expected output indicators, and unexpected output indicators. An efficiency frontier envelope is constructed based on a one-dimensional matrix, periodic indicator weights, a two-dimensional matrix, and the value-performance gap. The distance between this envelope and the expected standard efficiency frontier envelope is measured to determine the operational evaluation results. Based on the operational evaluation results, the first-order optimization objective is to coordinate the resource transfer between input and output combinations, and the second-order optimization objective is to improve the correlation between each input and output combination. The resulting combination improvement strategy is then determined. Integrate the operational evaluation results with combined improvement strategies to provide operational management alerts and optimization management guidance.

2. The dual-dimensional evaluation method for the operation and management of a water purification plant as described in claim 1, characterized in that, The weights of the periodic indicators are obtained, including: According to the preset cycle, periodic indicators are collected from the target water purification plant to obtain a one-dimensional data matrix. The periodic indicators include pollutant removal rate, electricity consumption per ton of water, and chemical consumption per ton of water. After normalizing the one-dimensional matrix, the index entropy value is calculated to determine the objective weight of the index. The smaller the entropy value, the higher the objective weight. The standard weight distribution is invoked, and entropy value adjustment is performed on the standard weight distribution according to the objective weight of the indicator to obtain the periodic indicator weight.

3. The dual-dimensional evaluation method for the operation and management of a water purification plant as described in claim 1, characterized in that, Identify the value-performance gap, including: For the region where the plant is located, determine the marginal value coefficient and incentive weight of each operating indicator. The operating indicators include COD reduction rate, ammonia nitrogen reduction rate and influent COD concentration adjustment coefficient. Based on a preset cycle, for the aforementioned operational indicators, a binary matrix of data is collected from each water purification plant to determine the marginal value curve and weight curve. Based on the marginal value curve and weight curve, the value-performance gap is identified, where the value-performance gap is the gap item with high weight and low score in the curve.

4. The dual-dimensional evaluation method for the operation and management of a water purification plant as described in claim 1, characterized in that, Determine the operational evaluation results, including: For the target water purification plant, based on input indicators, expected output indicators and unexpected output indicators, the input-output combination is defined as the vertex, and a first efficiency front envelope is constructed, wherein the vertex efficiency value of the first efficiency front envelope is always 1. Based on the one-dimensional matrix, periodic index weights, two-dimensional matrix, and value-performance gap, a second efficiency frontier envelope is constructed. The operational evaluation results are determined based on the first efficiency front envelope and the second efficiency front envelope.

5. The dual-dimensional evaluation method for the operation and management of a water purification plant as described in claim 4, characterized in that, Based on the aforementioned one-dimensional matrix, periodic index weights, two-dimensional matrix, and value-performance gap, a second efficiency frontier envelope is constructed, including: Based on the aforementioned one-dimensional matrix, periodic index weights, two-dimensional matrix, and value-performance gap, the vertex efficiency value is determined. Vertices with an efficiency value of 1 are located on the leading edge surface, and vertices with an efficiency value less than 1 are located inside the leading edge surface. Vertices located on the leading edge surface are designated as leading edge vertices. Adjacent leading edge vertices are connected by a linear envelope to form the second efficiency leading edge envelope surface.

6. The dual-dimensional evaluation method for the operation and management of a water purification plant as described in claim 5, characterized in that, For each defined input-output combination, the objective function is to maximize the weighted ratio of output to input as the efficiency value. The constraint is that the weighted sum of output of all input-output combinations does not exceed the weighted sum of input. Linear programming is used to solve the problem based on a one-dimensional matrix, periodic index weights, a two-dimensional matrix, and the value-performance gap to obtain the vertex efficiency value of each input-output combination.

7. The dual-dimensional evaluation method for the operation and management of a water purification plant as described in claim 6, characterized in that, Based on the first efficiency frontier envelope and the second efficiency frontier envelope, the operational evaluation results are determined, including: The mapping distance between the first efficiency front envelope and the second efficiency front envelope is calculated to obtain the distance vector from the internal vertex of each front to the front. The distance vector is based on a joint measure of input redundancy, expected output deficiency, and unexpected output excess. Based on each distance vector, the operational evaluation results are obtained.

8. The dual-dimensional evaluation method for the operation and management of a water purification plant as described in claim 7, characterized in that, Using the coordination of resource transfers among input-output combinations as the first-order optimization objective and the improvement of the correlation between each input-output combination as the second-order optimization objective, the resulting combination improvement strategies are determined, including: Based on the input redundancy, expected output shortfall, and unexpected output excess corresponding to each input-output combination, resource coordination is carried out with the resource transfer target between combinations to obtain the first resource coordination strategy. Based on the first resource coordination strategy, and based on the set of improvement strategies associated with each input-output combination, control decisions are made to obtain a combined improvement strategy.

9. The dual-dimensional evaluation method for the operation and management of a water purification plant as described in claim 8, characterized in that, Based on the combined improvement strategy, the second efficiency front envelope is updated, and it is verified whether the target is met. If there are vertices inside the front, iterative optimization is performed until convergence, and the combined improvement strategy is updated.

10. A two-dimensional evaluation system for the operation and management of a water purification plant, characterized in that, The system is used to implement the dual-dimensional evaluation method for the operation and management of a water purification plant as described in any one of claims 1-9, and the system includes: The periodic indicator weight acquisition module is used to collect a one-dimensional array of data from the target water purification plant at a preset period, and obtain the periodic indicator weights through normalized indicator entropy value analysis. The gap identification module is used to collect two-dimensional data arrays of the target water purification plant at preset intervals, and identify value-performance gaps based on the marginal value coefficients and incentive weights of various operating indicators in the region where the plant is located. The operation evaluation result determination module is used to define input-output combinations with input indicators, expected output indicators and unexpected output indicators, construct the efficiency frontier envelope based on a one-dimensional matrix, periodic indicator weights, a two-dimensional matrix and the value-performance gap, measure the distance with the expected standard efficiency frontier envelope, and determine the operation evaluation result. The combined improvement strategy acquisition module is used to determine the combined improvement strategy based on the operational evaluation results, with the resource transfer coordination between input-output combinations as the first-order optimization objective and the correlation improvement of each input-output combination as the second-order optimization objective. The operations management alarm module is used to integrate the operations evaluation results with combined improvement strategies to provide operations management alarms and optimization management guidance.

Images