Wastewater heavy metal recovery intelligent decision-making method for metal processing

Abstract

The invention relates to the technical field of data processing, and discloses a metal processing wastewater heavy metal recovery intelligent decision-making method, which comprises the following steps: extracting key feature vectors of normalized multi-dimensional physical and chemical parameters; dynamically matching the key feature vector with a preset process combination to obtain a process configuration scheme; constructing a first evaluation factor based on the economic indexes of the heavy metal recovery rate and the reagent consumption, and constructing a second evaluation factor based on the environmental protection indexes of the sludge output and the energy consumption; performing Pareto frontier analysis on the process configuration scheme based on the first evaluation factor and the second evaluation factor to dynamically obtain an optimization evaluation result of the process configuration scheme; generating a heavy metal recovery decision instruction of metal processing according to the optimization evaluation result, and triggering a start-stop control signal of equipment in metal processing based on the heavy metal recovery decision instruction; according to the invention, the efficiency of intelligent decision-making for wastewater heavy metal recovery can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a wastewater heavy metal recovery intelligent decision-making method for metal processing. BACKGROUND

[0002] In the field of wastewater heavy metal recovery in metal processing, the existing technology often has difficulty in efficiently integrating multi-dimensional physicochemical parameters of wastewater in the decision-making process, resulting in insufficient precision in grasping the characteristics of the wastewater, and thus affecting the rationality and pertinence of process matching. Due to the lack of effective extraction and analysis of key features, the generation of process configuration schemes is mostly dependent on experience or simple matching, which is difficult to adapt to complex and variable wastewater conditions, resulting in low decision-making efficiency and inability to quickly respond to actual processing needs.

[0003] At the same time, the existing technology often only considers economic or environmental indicators when evaluating process schemes, failing to achieve a comprehensive dynamic trade-off between the two. This results in the optimized process scheme often being one-sided, either economically unsatisfactory, increasing processing costs, or environmentally insufficient, not meeting green development requirements. Moreover, the evaluation process lacks a systematic analysis method, making it difficult to dynamically obtain the optimal process configuration scheme, further reducing the overall effectiveness of heavy metal recovery intelligent decision-making. SUMMARY

[0004] The present application provides a wastewater heavy metal recovery intelligent decision-making method for metal processing to solve the problems raised in the background art.

[0005] To achieve the above-mentioned purpose, the present application provides a wastewater heavy metal recovery intelligent decision-making method for metal processing, comprising: S1, obtaining multi-dimensional physicochemical parameters of wastewater samples in metal processing, and extracting key feature vectors of normalized multi-dimensional physicochemical parameters; S2, dynamically matching the key feature vectors with a preset process combination to obtain a process configuration scheme for the wastewater samples; S3, constructing a first evaluation factor of the process configuration scheme based on the economic indicators of heavy metal recovery rate and reagent consumption in the process configuration scheme; S4, constructing a second evaluation factor of the process configuration scheme based on the environmental indicators of sludge production and energy consumption in the process configuration scheme; S5, performing a Pareto frontier analysis on the process configuration scheme based on the first and second evaluation factors to dynamically obtain an optimized evaluation result of the process configuration scheme; S6, generating a heavy metal recovery decision-making instruction for the metal processing according to the optimized evaluation result, and triggering a start-stop control signal of the equipment in the metal processing based on the heavy metal recovery decision-making instruction.

[0006] In a preferred embodiment, the step of obtaining multi-dimensional physicochemical parameters of wastewater samples from metal processing and extracting key feature vectors of the normalized multi-dimensional physicochemical parameters includes: Obtain the chemical and physical properties of wastewater samples from metal processing to obtain the original parameter set of the wastewater samples; The original parameter set is subjected to a standardization transformation to obtain the normalized parameter matrix of the wastewater sample; Separate the heavy metal concentration characteristic group and the reaction environment characteristic group from the normalized parameter matrix; Principal component dimensionality reduction is performed on the heavy metal concentration feature group to obtain the core concentration vector of the wastewater sample. The core concentration vector is concatenated with the reaction environment feature group to obtain the key feature vector of the wastewater sample.

[0007] In a preferred embodiment, the step of dynamically matching the key feature vector with a preset process combination to obtain a process configuration scheme for the wastewater sample includes: The heavy metal concentration feature group in the key feature vector is input into a preset process combination library for primary matching to obtain the candidate process set of the wastewater sample. Based on the reaction environment feature group in the key feature vector, the candidate process set is filtered for environmental adaptability to obtain the qualified process node of the wastewater sample. The qualified process nodes are topologically sorted to obtain the process configuration scheme for the wastewater sample.

[0008] In a preferred embodiment, the step of performing topological sorting on the qualified process nodes to obtain the process configuration scheme for the wastewater sample includes: The qualified process nodes are topologically sorted to obtain the initial combination framework of the wastewater sample; Obtain the process execution record from historical wastewater treatment data that has the highest similarity to the key feature vector; Extract the parameter correction items from the process execution record, and dynamically calibrate the initial combined frame based on the parameter correction items; When conflicts exist between process nodes in the initial combined framework after dynamic calibration, a priority arbitration mechanism is triggered to obtain the process configuration scheme of the wastewater sample.

[0009] In a preferred embodiment, the first evaluation factor is calculated using the following formula: In the formula, The first evaluation factor, This is the recovery rate weighting coefficient. The target recovery rate of heavy metals extracted in the aforementioned process configuration scheme. The baseline recovery rate is defined in the historical wastewater treatment data. This refers to the baseline reagent consumption in the historical wastewater treatment data. This refers to the unit consumption of reagents in the process configuration scheme.

[0010] In a preferred embodiment, the second evaluation factor is calculated using the following formula: In the formula, This is the second evaluation factor. This is the sludge weighting coefficient. This serves as the baseline sludge volume in the environmental standards database. The benchmark energy consumption in the environmental protection standard library, The predicted sludge production extracted from the aforementioned process configuration scheme, The total energy consumption extracted from the process configuration scheme.

[0011] In a preferred embodiment, the step of dynamically obtaining the optimization evaluation result of the process configuration scheme by performing Pareto front analysis based on the first evaluation factor and the second evaluation factor includes: Using the first evaluation factor as the economic coordinate axis and the second evaluation factor as the environmental coordinate axis, the historical process schemes are mapped to a set of discrete points in the solution space to obtain the multi-objective solution space of the process configuration scheme; By traversing the discrete points of the multi-objective solution space, the Pareto front candidate set of the process configuration scheme is obtained; The implementation preference weight vector of the process configuration scheme is generated based on the weight factors of the first evaluation factor and the second evaluation factor. The implementation preference weight vector is projected onto the Pareto front candidate set to obtain the optimization evaluation result of the process configuration scheme.

[0012] In a preferred embodiment, the formula for calculating the preference weight vector is as follows: In the formula, For the implementation preference weight vector, For vector normalization operation, This is the economic weight component in the first evaluation factor. As a market sensitivity adjustment factor, This is the environmental weight component in the second evaluation factor. The policy reinforcement index.

[0013] In a preferred embodiment, generating the heavy metal recycling decision instruction for metal processing based on the optimization evaluation result includes: Analyze the process parameter configurations in the optimization evaluation results; The process parameter configuration is converted into a sequence of executable instructions for the equipment; The executable instruction sequence of the device is subjected to security verification; By integrating the sequence of executable instructions from the equipment that has passed security verification, the heavy metal recycling decision instructions for the metal processing are obtained.

[0014] In a preferred embodiment, triggering the start / stop control signal of the metal processing equipment based on the heavy metal recycling decision command includes: The process stage sequence in the heavy metal recycling decision instruction is analyzed to obtain the time-series control node set of the metal processing; The device start / stop logic is mapped according to the timing control node set; The start / stop control signal of the metal processing equipment is triggered according to the equipment start / stop logic.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention achieves accurate capture of wastewater sample characteristics by extracting key feature vectors of normalized multi-dimensional physicochemical parameters. Combined with a dynamic matching mechanism and corresponding preset process combinations, it can quickly generate suitable process configuration schemes, effectively improving the accuracy and efficiency of process matching, providing a reliable basis for subsequent decision-making, and thus improving the overall efficiency of intelligent decision-making for wastewater heavy metal recovery.

[0016] 2. This invention constructs a first evaluation factor and a second evaluation factor, comprehensively considering economic and environmental indicators, and dynamically obtains optimized evaluation results through Pareto front analysis. The resulting decision commands can accurately trigger equipment start-up and shutdown control signals. This process achieves comprehensive optimization of the process configuration scheme, ensuring both the economic efficiency and environmental friendliness of heavy metal recovery, thereby improving the overall effect of heavy metal recovery and treatment of metal processing wastewater. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating an intelligent decision-making method for heavy metal recovery from wastewater in metal processing, provided in an embodiment of the present invention. The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides an intelligent decision-making method for heavy metal recovery from wastewater in metal processing. The executing entity of this intelligent decision-making method for heavy metal recovery from wastewater in metal processing includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the intelligent decision-making method for heavy metal recovery from wastewater in metal processing can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0020] Reference Figure 1 The diagram shown is a flowchart illustrating an intelligent decision-making method for heavy metal recovery from wastewater in metal processing, provided by an embodiment of the present invention. In this embodiment, the intelligent decision-making method for heavy metal recovery from wastewater in metal processing includes: S1. Obtain the multi-dimensional physicochemical parameters of wastewater samples from metal processing, and extract the key feature vectors of the normalized multi-dimensional physicochemical parameters. In this embodiment of the invention, the step of obtaining multi-dimensional physicochemical parameters of wastewater samples from metal processing and extracting key feature vectors of the normalized multi-dimensional physicochemical parameters includes: Obtain the chemical and physical properties of wastewater samples from metal processing to obtain the original parameter set of the wastewater samples; The original parameter set is subjected to a standardization transformation to obtain the normalized parameter matrix of the wastewater sample; Separate the heavy metal concentration characteristic group and the reaction environment characteristic group from the normalized parameter matrix; Principal component dimensionality reduction is performed on the heavy metal concentration feature group to obtain the core concentration vector of the wastewater sample. The core concentration vector is concatenated with the reaction environment feature group to obtain the key feature vector of the wastewater sample.

[0021] Specifically, to obtain the raw parameter set for a metal processing wastewater sample, it is necessary to first determine the chemical and physical parameters to be tested. Chemical parameters include the concentration of various heavy metals (such as lead, mercury, and chromium), pH value, chemical oxygen demand (COD), biochemical oxygen demand (BOD), chloride ion content, and sulfate ion content. Physical parameters include the wastewater's temperature, color, transparency, turbidity, suspended solids content, and odor. Then, a certain amount of metal processing wastewater sample is collected, and these parameters are measured using appropriate testing tools and methods. For example, pH is directly measured using a pH meter, the concentration of various heavy metals is determined using an atomic absorption spectrophotometer, temperature is measured using a thermometer, and color and transparency are detected through observation and specific instruments. All the measured data are then compiled and summarized to obtain the raw parameter set for the wastewater sample.

[0022] Furthermore, when performing standardization transformation on the original parameter set, the value range of each parameter is first defined. For each parameter, the minimum value of the parameter in the original parameter set is subtracted from each specific value of the parameter, and the difference is then divided by the difference between the maximum and minimum values ​​of the parameter. Through this calculation, each data in the original parameter set is transformed to the range of 0 to 1, and the resulting matrix is ​​the normalized parameter matrix of the wastewater sample.

[0023] Furthermore, to separate the heavy metal concentration characteristic group and the reaction environment characteristic group from the normalized parameter matrix, it is necessary to first determine which parameters belong to the heavy metal concentration characteristics and which belong to the reaction environment characteristics. The heavy metal concentration characteristic group contains the concentration data of various heavy metals (such as lead, mercury, chromium, etc.) in the wastewater, while the reaction environment characteristic group contains data such as pH value, temperature, chemical oxygen demand, biochemical oxygen demand, chloride ion content, sulfate ion content, transparency, turbidity, suspended solids content, and odor. Then, all parameters belonging to the heavy metal concentration characteristics are extracted from the normalized parameter matrix to form the heavy metal concentration characteristic group; all parameters belonging to the reaction environment characteristics are extracted to form the reaction environment characteristic group.

[0024] Furthermore, principal component analysis (PCA) is performed on the heavy metal concentration feature set. First, the correlation between the various heavy metal concentration parameters in the feature set is analyzed. A few comprehensive indicators that reflect the main information of these parameters are identified; these comprehensive indicators are the principal components. Through calculation, the original multiple heavy metal concentration parameters are transformed into these principal components. This ensures that the newly obtained principal components retain the information from the original heavy metal concentration feature set to the greatest extent possible while reducing the number of parameters. The resulting vector is the core concentration vector of the wastewater sample.

[0025] Furthermore, the core concentration vector is concatenated with the reaction environment feature set. This involves arranging and combining the elements of the core concentration vector with the parameters of the reaction environment feature set in a specific order to form a new vector. This new vector contains both the core information of heavy metal concentration after dimensionality reduction and the feature information of the reaction environment, which is the key feature vector of the wastewater sample.

[0026] In summary, by obtaining chemical and physical property parameters to form an initial parameter set, and then standardizing and transforming it to obtain a normalized parameter matrix, the influence of dimensional differences between different parameters can be eliminated, ensuring the comparability of parameters and laying a unified foundation for subsequent characteristic analysis.

[0027] In summary, after separating the heavy metal concentration feature group from the reaction environment feature group, principal component analysis was performed on the heavy metal concentration feature group to obtain the core concentration vector, which was then concatenated with the reaction environment feature group to form the key feature vector. This not only preserved the core information related to heavy metal recovery in the wastewater sample, but also achieved reasonable simplification of feature dimensions, avoiding interference from redundant information. This makes the subsequent dynamic matching with the preset process combination more accurate and efficient, providing strong support for the scientific nature of the process configuration scheme.

[0028] S2. Dynamically match the key feature vectors with the preset process combinations to obtain the process configuration scheme of the wastewater sample; In this embodiment of the invention, the step of dynamically matching the key feature vector with a preset process combination to obtain the process configuration scheme for the wastewater sample includes: The heavy metal concentration feature group in the key feature vector is input into a preset process combination library for primary matching to obtain the candidate process set of the wastewater sample. Based on the reaction environment feature group in the key feature vector, the candidate process set is filtered for environmental adaptability to obtain the qualified process node of the wastewater sample. The qualified process nodes are topologically sorted to obtain the process configuration scheme for the wastewater sample.

[0029] The process configuration scheme for the wastewater sample is obtained by performing topological sorting on the qualified process nodes, including: The qualified process nodes are topologically sorted to obtain the initial combination framework of the wastewater sample; Obtain the process execution record from historical wastewater treatment data that has the highest similarity to the key feature vector; Extract the parameter correction items from the process execution record, and dynamically calibrate the initial combined frame based on the parameter correction items; When conflicts exist between process nodes in the initial combined framework after dynamic calibration, a priority arbitration mechanism is triggered to obtain the process configuration scheme of the wastewater sample.

[0030] Specifically, heavy metal concentration feature groups are extracted from key feature vectors and compared with a pre-defined process combination library. This library contains various process combinations targeting different heavy metal concentration characteristics, each corresponding to a specific heavy metal concentration range and characteristic performance. During the comparison, it is checked whether each data point in the heavy metal concentration feature group meets the applicable conditions of a particular process combination in the library. All process combinations that meet the conditions are selected, and these selected process combinations together constitute the candidate process set for the wastewater sample.

[0031] Furthermore, based on the reaction environment feature group in the key feature vector, environmental adaptability filtering is performed on the candidate process set. First, the influence requirements of each parameter in the reaction environment feature group (such as pH, temperature, etc.) on each process in the candidate process set are clarified. Each process in the candidate process set is examined one by one to determine whether it can operate normally under the environmental conditions presented by the reaction environment feature group. For example, if a process requires operation within a specific pH range, and the pH value in the reaction environment feature group is not within that range, the process is excluded. After this screening, the remaining process nodes are the qualified process nodes for the wastewater sample.

[0032] Furthermore, a topological sort is performed on the qualified process nodes. First, the sequential dependencies between the qualified process nodes are analyzed to determine which process nodes must be executed before other process nodes and which process nodes need to be executed after other process nodes. Following this sequential order, all qualified process nodes are arranged in sequence to form an ordered sequence. This sequence is the process configuration scheme for the wastewater sample.

[0033] Furthermore, a topological sort is performed on the qualified process nodes. First, the prerequisite and successor relationships between each qualified process node are clarified, that is, it is determined that a certain process node can only begin after another process node is completed, or that the completion of one process node is a prerequisite for the start of another process node. Based on these relationships, starting with process nodes without prerequisite dependencies, the positions of each process node are arranged sequentially, ensuring that each process node appears after all its preceding process nodes. The resulting ordered arrangement forms the initial composite framework for the wastewater sample.

[0034] Furthermore, the process execution record with the highest similarity to the key feature vector in the historical wastewater treatment data is obtained. First, each data point in the key feature vector is compared item by item with the feature vector corresponding to each process execution record in the historical wastewater treatment data. During the comparison, the degree of similarity between each data point is considered, and the overall consistency between each historical record and the key feature vector is calculated. The process execution record with the highest consistency is selected; this record is the required process execution record.

[0035] Furthermore, parameter correction items are extracted from the process execution record. These items are the specific adjustments made to the operating parameters of the process nodes, such as extending the reaction time or increasing or decreasing the dosage of reagents. Based on these parameter correction items, the initial assembly framework is dynamically calibrated. That is, according to the requirements of the correction items, the operating parameters of the corresponding process nodes in the initial assembly framework are adjusted to ensure that the adjusted process node parameters are consistent with the correction items, thus completing the dynamic calibration of the initial assembly framework.

[0036] Furthermore, when conflicts exist between process nodes in the initial combined framework after dynamic calibration, these conflicts manifest as situations where two or more process nodes cannot simultaneously meet certain requirements regarding operating conditions and timing. A priority arbitration mechanism is triggered. This mechanism pre-sets the priority order of different process nodes. Based on this order, the operating requirements of higher-priority process nodes are retained first, while the operating requirements of lower-priority process nodes are adjusted to eliminate conflicts. The resulting ordered sequence of process nodes after adjustment constitutes the process configuration scheme for the wastewater sample.

[0037] In summary, by inputting the heavy metal concentration feature group from the key feature vector into a preset process combination library for initial matching, a set of candidate processes that are compatible with the heavy metal concentration characteristics of wastewater can be quickly screened, providing a targeted basis for subsequent process configuration. Furthermore, by combining the reaction environment feature group to perform environmental adaptability filtering on the candidate process set, it can be further ensured that the selected qualified process nodes can adapt to the reaction environment conditions of wastewater, thereby improving the accuracy of process matching.

[0038] In summary, after obtaining the initial combined framework by topological sorting of qualified process nodes, dynamic calibration is performed by combining parameter correction items in historical similar process execution records, and a priority arbitration mechanism is triggered when conflicts exist. This enables the process configuration scheme to not only conform to the actual characteristics of wastewater, but also to achieve optimization and adjustment by drawing on historical experience. The final process configuration scheme is more reasonable and feasible, providing a reliable process basis for subsequent evaluation and decision-making, and improving the overall decision-making efficiency.

[0039] S3. Based on the economic indicators of heavy metal recovery rate and reagent consumption in the process configuration scheme, construct the first evaluation factor of the process configuration scheme; In this embodiment of the invention, the calculation formula for the first evaluation factor is as follows: In the formula, The first evaluation factor, This is the recovery rate weighting coefficient. The target recovery rate of heavy metals extracted in the aforementioned process configuration scheme. The baseline recovery rate is defined in the historical wastewater treatment data. This refers to the baseline reagent consumption in the historical wastewater treatment data. This refers to the unit consumption of reagents in the process configuration scheme.

[0040] Specifically, each component of the first evaluation factor has a clear source: the recovery rate weighting coefficient is a pre-set value based on the consideration of the importance of heavy metal recovery in actual applications; the heavy metal target recovery rate comes from the heavy metal recovery target ratio specified in the process configuration plan; the baseline recovery rate is the conventional recovery ratio statistically derived from historical wastewater treatment data; the baseline reagent consumption is the conventional quantity of reagents consumed per unit of wastewater treated in historical wastewater treatment data; and the reagent unit consumption is the quantity of reagents planned to be consumed per unit of wastewater treated in the process configuration plan.

[0041] Furthermore, the significance of this formula lies in comprehensively evaluating the overall benefits of the process configuration scheme in terms of heavy metal recovery and reagent consumption. By weighting and summing the ratio of the target recovery rate of heavy metals to the benchmark recovery rate and the ratio of the benchmark reagent consumption to the unit reagent consumption according to the recovery rate weighting coefficient and their complementary coefficient, the first evaluation factor can intuitively reflect the comprehensive performance of the process configuration scheme in terms of economy and ecology compared with the historical conventional treatment methods.

[0042] Furthermore, the formula shows that when the target recovery rate of heavy metals is higher and the ratio to the benchmark recovery rate is larger, the first evaluation factor will increase accordingly; when the reagent unit consumption is lower and the ratio of the benchmark reagent consumption to the reagent unit consumption is larger, the first evaluation factor will also increase; conversely, if the target recovery rate of heavy metals decreases or the reagent unit consumption increases, the first evaluation factor will decrease, and the magnitude of the recovery rate weighting coefficient will affect the degree of influence of the former term on the first evaluation factor, while its complementary coefficient will affect the degree of influence of the latter term.

[0043] In summary, this evaluation factor, by comparing the target heavy metal recovery rate of the process configuration with historical baseline recovery rates and the unit reagent consumption with historical baseline reagent consumption, and combining this with a recovery rate weighting coefficient, can quantitatively reflect the economic performance of the process configuration. This quantitative approach allows for direct comparison of the economic differences between different process configurations, providing a clear economic benchmark for subsequent optimization.

[0044] In summary, by integrating the two core economic indicators of recovery rate and reagent consumption, the first evaluation factor focuses on both the efficiency of heavy metal recovery (recovery rate) and the cost input in the treatment process (reagent consumption), achieving a comprehensive assessment of the economics of process configuration schemes. This helps to prioritize process schemes with high recovery rates and low reagent consumption during the decision-making process, improving resource recovery efficiency while reducing treatment costs, and providing strong support for achieving cost-effective and efficient heavy metal recovery.

[0045] S4. Based on the environmental protection indicators of sludge production and energy consumption in the process configuration scheme, construct a second evaluation factor for the process configuration scheme; In this embodiment of the invention, the calculation formula for the second evaluation factor is as follows: In the formula, This is the second evaluation factor. This is the sludge weighting coefficient. This serves as the baseline sludge volume in the environmental standards database. The benchmark energy consumption in the environmental protection standard library, The predicted sludge production extracted from the aforementioned process configuration scheme, The total energy consumption extracted from the process configuration scheme.

[0046] Specifically, the sludge weighting coefficient is a pre-set value based on the importance attached to sludge treatment in the actual scenario; the baseline sludge quantity comes from the standard quantity of sludge that can be generated by a unit of wastewater treatment as specified in the environmental protection standard library; the baseline energy consumption is the standard quantity of energy that can be consumed by a unit of wastewater treatment as specified in the environmental protection standard library; the predicted sludge generation is extracted from the process configuration scheme, which predicts the amount of sludge that will be generated by a unit of wastewater treatment; and the comprehensive energy consumption is extracted from the process configuration scheme, which is the sum of all types of energy required to treat a unit of wastewater.

[0047] Furthermore, the significance of this formula lies in comprehensively evaluating the environmental performance of the process configuration scheme in terms of sludge generation and energy consumption. By weighting and summing the ratio of baseline sludge quantity to predicted sludge generation and the ratio of baseline energy consumption to comprehensive energy consumption according to the sludge weight coefficient and its complementary coefficient, the resulting second evaluation factor can intuitively reflect the comprehensive performance of the process configuration scheme in terms of ecological and environmental protection compared to environmental protection standards.

[0048] Furthermore, the formula shows that when the predicted sludge production is less and the ratio of baseline sludge production to predicted sludge production is larger, the second assessment factor will increase accordingly; when the comprehensive energy consumption is lower and the ratio of baseline energy consumption to comprehensive energy consumption is larger, the second assessment factor will also increase; conversely, if the predicted sludge production increases or comprehensive energy consumption increases, the second assessment factor will decrease, and the sludge weight coefficient will affect the degree of influence of the former on the second assessment factor, while its complementarity coefficient will affect the degree of influence of the latter.

[0049] In summary, this assessment factor, by comparing the predicted sludge production of the process configuration scheme with the benchmark sludge production in the environmental protection standard library, and the comprehensive energy consumption with the benchmark energy consumption in the environmental protection standard library, combined with the sludge weighting coefficient, can intuitively reflect the environmental performance of the process scheme, provide a comparable measurement scale for the differences in environmental performance of different process configuration schemes, and make environmental assessment more objective and scientific.

[0050] In summary, by integrating the two key environmental indicators of sludge generation and energy consumption, the second evaluation factor considers both the secondary generation of pollutants (sludge) and the rational use of energy (energy consumption) during the treatment process, achieving a comprehensive assessment of the environmental friendliness of the process configuration. This helps to prioritize process schemes with low sludge generation and low energy consumption during the decision-making process, ensuring the effectiveness of heavy metal recovery while reducing negative environmental impacts, and providing strong support for achieving green and environmentally friendly heavy metal recovery and treatment.

[0051] S5. Based on the first evaluation factor and the second evaluation factor, Pareto front analysis is performed on the process configuration scheme to dynamically obtain the optimization evaluation result of the process configuration scheme; In this embodiment of the invention, the step of dynamically obtaining the optimization evaluation result of the process configuration scheme by performing Pareto front analysis on the process configuration scheme based on the first evaluation factor and the second evaluation factor includes: Using the first evaluation factor as the economic coordinate axis and the second evaluation factor as the environmental coordinate axis, the historical process schemes are mapped to a set of discrete points in the solution space to obtain the multi-objective solution space of the process configuration scheme; By traversing the discrete points of the multi-objective solution space, the Pareto front candidate set of the process configuration scheme is obtained; The implementation preference weight vector of the process configuration scheme is generated based on the weight factors of the first evaluation factor and the second evaluation factor. The implementation preference weight vector is projected onto the Pareto front candidate set to obtain the optimization evaluation result of the process configuration scheme.

[0052] The formula for calculating the implementation preference weight vector is as follows: In the formula, For the implementation preference weight vector, For vector normalization operation, This is the economic weight component in the first evaluation factor. As a market sensitivity adjustment factor, This is the environmental weight component in the second evaluation factor. The policy reinforcement index.

[0053] Specifically, a two-dimensional coordinate system is defined, where the horizontal axis represents economic efficiency, expressed by the value of the first evaluation factor, and the vertical axis represents environmental friendliness, expressed by the value of the second evaluation factor. This coordinate system constitutes the solution space of the process configuration scheme. Historical process schemes are collected, and the first and second evaluation factors corresponding to each historical process scheme are calculated. Then, in this two-dimensional coordinate system, a point is determined for each historical process scheme according to its corresponding first and second evaluation factor values. The set of all these points together constitutes the multi-objective solution space of the process configuration scheme.

[0054] Furthermore, each discrete point in the multi-objective solution space is traversed, and the first evaluation factor and the second evaluation factor are compared one by one. For a point, if there is no other point whose first evaluation factor is not less than its own and whose second evaluation factor is also greater than its own, then this point is selected into the Pareto front candidate set. In this way, all points that meet the conditions are selected, and these points together constitute the Pareto front candidate set of this process configuration scheme.

[0055] Furthermore, based on the emphasis on economy and environmental protection in actual needs, weighting factors for the first evaluation factor and the second evaluation factor are set, with the sum of the two weighting factors being 1. The weighting factors of the first evaluation factor and the second evaluation factor are combined in sequence to form a vector, which is the implementation preference weight vector of the process configuration scheme.

[0056] Furthermore, the direction represented by the implementation preference weight vector is projected onto the Pareto front candidate set, and the point furthest from this projection direction is found. The combination of the first evaluation factor and the second evaluation factor corresponding to this point is the optimized evaluation result of the process configuration scheme.

[0057] Specifically, the economic weighting component is a pre-set value based on the degree of importance attached to the economic aspects represented by the first evaluation factor in the actual scenario; the market sensitivity adjustment factor is determined based on the current market's sensitivity to economic changes, and the more the market focuses on economic changes, the higher this factor's value; the environmental weighting component is a pre-set value based on the degree of importance attached to the environmental aspects represented by the second evaluation factor in the actual scenario; and the policy reinforcement index is determined based on the strictness and enforcement of current environmental policies, and the stricter the policies, the higher this index's value.

[0058] Furthermore, the significance of this formula lies in obtaining two product results by multiplying the economic weight component with the market sensitivity adjustment factor and the environmental weight component with the policy reinforcement index. Then, the vector formed by these two results is normalized to make the length of the vector 1. The final implementation preference weight vector can comprehensively reflect the market's sensitivity to economic factors and the degree of policy reinforcement of environmental factors, thereby more accurately reflecting the actual preference for economic and environmental factors.

[0059] Furthermore, vector normalization is achieved by calculating the square root of the sum of the squares of the two product results, and then dividing each product result by this square root to obtain a vector with a length of 1. The formula shows the following trend: when the market sensitivity adjustment factor increases, the product of the economic weight component and this factor increases; after normalization, the proportion of the economic component in the implementation preference weight vector will increase. When the policy strengthening index increases, the product of the environmental weight component and this index increases; after normalization, the proportion of the environmental component in the implementation preference weight vector will increase. Conversely, if the market sensitivity adjustment factor decreases, the proportion of the economic component decreases, the policy strengthening index decreases, and the proportion of the environmental component decreases.

[0060] In summary, by using the first evaluation factor as the economic axis and the second evaluation factor as the environmental axis, historical process schemes are mapped to a discrete set of points in the solution space, constructing a multi-objective solution space. This clearly presents the distribution characteristics of different process schemes in terms of economics and environmental protection, providing an intuitive visualization basis for subsequent analysis. Traversing this solution space yields a Pareto front candidate set, which can screen out the optimal set of schemes that cannot simultaneously optimize economics and environmental protection under current conditions, ensuring the scientific validity of the evaluation results.

[0061] In summary, by combining the weights of the first and second evaluation factors to generate an implementation preference weight vector, and projecting this vector onto the Pareto frontier candidate set, the optimized evaluation results are obtained. This approach considers both economic and environmental balance, and allows for dynamic adjustments based on actual market sensitivity and policy requirements, making the optimized results more aligned with real-world application scenarios. This dynamic analysis method can accurately capture the optimal balance point of process configuration schemes, providing a reliable basis for generating reasonable recycling decision instructions and effectively improving the accuracy and adaptability of decision-making.

[0062] S6. Generate a heavy metal recycling decision instruction for the metal processing based on the optimization evaluation results, and trigger the start / stop control signal of the equipment in the metal processing based on the heavy metal recycling decision instruction.

[0063] In this embodiment of the invention, generating the heavy metal recycling decision instruction for metal processing based on the optimization evaluation result includes: Analyze the process parameter configurations in the optimization evaluation results; The process parameter configuration is converted into a sequence of executable instructions for the equipment; The executable instruction sequence of the device is subjected to security verification; By integrating the sequence of executable instructions from the equipment that has passed security verification, the heavy metal recycling decision instructions for the metal processing are obtained.

[0064] The triggering of the start / stop control signal for the metal processing equipment based on the heavy metal recycling decision command includes: The process stage sequence in the heavy metal recycling decision instruction is analyzed to obtain the time-series control node set of the metal processing; The device start / stop logic is mapped according to the timing control node set; The start / stop control signal of the metal processing equipment is triggered according to the equipment start / stop logic.

[0065] Specifically, analyzing the process parameter configuration in the optimization evaluation results requires clarifying the various process operation parameters included in the optimization evaluation results. These parameters involve the specific operational requirements of each process node in the heavy metal recovery process, such as reaction time, reagent addition amount, temperature control range, etc. By extracting and clarifying the specific values ​​and applicable scenarios of each parameter one by one, the analysis of the process parameter configuration can be completed.

[0066] Furthermore, to convert process parameter configurations into a sequence of executable instructions for the equipment, it is necessary to first determine the equipment operation corresponding to each process parameter. For example, the reaction time corresponds to the equipment's running time setting, and the amount of reagent added corresponds to the opening degree and time of the dosing equipment. Then, according to the sequence of process nodes, these equipment operations are converted into specific instructions that the equipment can recognize, such as "Equipment A is turned on and will be turned off after running for 30 minutes" or "Equipment B is turned on after Equipment A is turned off, and will be turned off after adding 50 liters of reagent." These sequentially arranged instructions together constitute the sequence of executable instructions for the equipment.

[0067] Furthermore, a safety verification is performed on the sequence of executable instructions of the equipment. It is necessary to check whether each instruction complies with the equipment's safe operation specifications. For example, whether the equipment's operating time in the instruction is within the equipment's allowed continuous operating range, whether the amount of reagent added exceeds the equipment's maximum capacity, and whether there are safety hazards caused by conflicts between instructions from different equipment. If an instruction is found to be non-compliant with safety specifications, the instruction is corrected until all instructions pass the safety check.

[0068] Furthermore, to integrate the sequence of executable instructions from equipment that has passed safety verification, all verified instructions need to be organized in the order of process nodes to ensure smooth connection between instructions and that there are no duplicates or omissions. The final complete and orderly set of instructions is the heavy metal recycling decision instruction for metal processing.

[0069] Furthermore, based on the device start-stop logic mapped from the timing control node set, it is necessary to clarify the time point and associated device corresponding to each node in the timing control node set, and determine which devices need to start and which devices need to stop at each time point. For example, at the first time point, device C starts and device D remains off; at the second time point, device C stops and device E starts, etc. Through this correspondence, the logical rules for device start-stop are constructed.

[0070] Furthermore, based on the equipment start-stop logic, the start-stop control signals of the metal processing equipment are triggered. According to the time points and equipment operations specified in the equipment start-stop logic, start or stop signals are sent to the corresponding equipment at the corresponding time. For example, when the start time node of equipment C is reached, a start signal is sent to equipment C, and equipment C starts running after receiving the signal; when the stop time node of equipment C is reached, a stop signal is sent to equipment C, and equipment C stops running after receiving the signal, thereby realizing the control of equipment start-stop.

[0071] In summary, by analyzing and optimizing the process parameter configurations in the evaluation results, converting them into a sequence of executable instructions for the equipment, and performing safety verification, it is possible to ensure that the decision-making instructions are compatible with the actual equipment operation requirements and comply with safety standards, thereby avoiding equipment failures or safety risks caused by instruction errors and providing a reliable guarantee for the stable execution of the process.

[0072] In summary, by parsing the process stage sequence in the decision-making instructions to obtain the set of timing control nodes, and then mapping the equipment start-up and shutdown logic to trigger control signals, a precise connection from decision-making to equipment operation is achieved. This automated instruction conversion and execution mechanism reduces manual intervention, improves the timeliness and accuracy of equipment control, ensures the efficient implementation of process configuration schemes, and ultimately enhances the overall efficiency of heavy metal recovery from metal processing wastewater.

[0073] In the several embodiments provided by this invention, it should be understood that the disclosed method can be implemented in other ways.

[0074] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0075] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, and technology that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0076] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A smart decision-making method for heavy metal recovery from wastewater in metal processing, characterized in that, The method includes: S1. Obtain the multi-dimensional physicochemical parameters of wastewater samples from metal processing, and extract the key feature vectors of the normalized multi-dimensional physicochemical parameters. S2. Dynamically match the key feature vectors with the preset process combinations to obtain the process configuration scheme of the wastewater sample; S3. Based on the economic indicators of heavy metal recovery rate and reagent consumption in the process configuration scheme, construct the first evaluation factor of the process configuration scheme; S4. Based on the environmental protection indicators of sludge production and energy consumption in the process configuration scheme, construct a second evaluation factor for the process configuration scheme; S5. Based on the first evaluation factor and the second evaluation factor, Pareto front analysis is performed on the process configuration scheme to dynamically obtain the optimization evaluation result of the process configuration scheme; S6. Generate a heavy metal recycling decision instruction for the metal processing based on the optimization evaluation results, and trigger the start / stop control signal of the equipment in the metal processing based on the heavy metal recycling decision instruction.

2. The intelligent decision-making method for heavy metal recovery from wastewater in metal processing as described in claim 1, characterized in that, The process of obtaining multi-dimensional physicochemical parameters of wastewater samples from metal processing and extracting key feature vectors of the normalized multi-dimensional physicochemical parameters includes: Obtain the chemical and physical properties of wastewater samples from metal processing to obtain the original parameter set of the wastewater samples; The original parameter set is subjected to a standardization transformation to obtain the normalized parameter matrix of the wastewater sample; Separate the heavy metal concentration characteristic group and the reaction environment characteristic group from the normalized parameter matrix; Principal component dimensionality reduction is performed on the heavy metal concentration feature group to obtain the core concentration vector of the wastewater sample. The core concentration vector is concatenated with the reaction environment feature group to obtain the key feature vector of the wastewater sample.

3. The intelligent decision-making method for heavy metal recovery from wastewater in metal processing as described in claim 2, characterized in that, The step of dynamically matching the key feature vector with a preset process combination to obtain the process configuration scheme for the wastewater sample includes: The heavy metal concentration feature group in the key feature vector is input into a preset process combination library for primary matching to obtain the candidate process set of the wastewater sample. Based on the reaction environment feature group in the key feature vector, the candidate process set is filtered for environmental adaptability to obtain the qualified process node of the wastewater sample. The qualified process nodes are topologically sorted to obtain the process configuration scheme for the wastewater sample.

4. The intelligent decision-making method for heavy metal recovery from wastewater in metal processing as described in claim 3, characterized in that, The process configuration scheme for the wastewater sample is obtained by performing topological sorting on the qualified process nodes, including: The qualified process nodes are topologically sorted to obtain the initial combination framework of the wastewater sample; Obtain the process execution record from historical wastewater treatment data that has the highest similarity to the key feature vector; Extract the parameter correction items from the process execution record, and dynamically calibrate the initial combined frame based on the parameter correction items; When conflicts exist between process nodes in the initial combined framework after dynamic calibration, a priority arbitration mechanism is triggered to obtain the process configuration scheme of the wastewater sample.

5. The intelligent decision-making method for heavy metal recovery from wastewater in metal processing as described in claim 4, characterized in that, The formula for calculating the first evaluation factor is as follows: In the formula, The first evaluation factor, This is the recovery rate weighting coefficient. The target recovery rate of heavy metals extracted in the aforementioned process configuration scheme. The baseline recovery rate is defined in the historical wastewater treatment data. This refers to the baseline reagent consumption in the historical wastewater treatment data. This refers to the unit consumption of reagents in the process configuration scheme.

6. The intelligent decision-making method for heavy metal recovery from wastewater in metal processing as described in claim 1, characterized in that, The formula for calculating the second evaluation factor is as follows: In the formula, This is the second evaluation factor. This is the sludge weighting coefficient. This serves as the baseline sludge volume in the environmental standards database. The benchmark energy consumption in the environmental protection standard library, The predicted sludge production extracted from the aforementioned process configuration scheme, The total energy consumption extracted from the process configuration scheme.

7. The intelligent decision-making method for heavy metal recovery from wastewater in metal processing as described in claim 1, characterized in that, The process configuration scheme optimization evaluation result is obtained dynamically by performing Pareto front analysis on the process configuration scheme based on the first evaluation factor and the second evaluation factor, including: Using the first evaluation factor as the economic coordinate axis and the second evaluation factor as the environmental coordinate axis, the historical process schemes are mapped to a set of discrete points in the solution space to obtain the multi-objective solution space of the process configuration scheme; By traversing the discrete points of the multi-objective solution space, the Pareto front candidate set of the process configuration scheme is obtained; The implementation preference weight vector of the process configuration scheme is generated based on the weight factors of the first evaluation factor and the second evaluation factor. The implementation preference weight vector is projected onto the Pareto front candidate set to obtain the optimization evaluation result of the process configuration scheme.

8. The intelligent decision-making method for heavy metal recovery from wastewater in metal processing as described in claim 7, characterized in that, The formula for calculating the implementation preference weight vector is as follows: In the formula, For the implementation preference weight vector, For vector normalization operation, This is the economic weight component in the first evaluation factor. As a market sensitivity adjustment factor, This is the environmental weight component in the second evaluation factor. The policy reinforcement index.

9. The intelligent decision-making method for heavy metal recovery from wastewater in metal processing as described in claim 1, characterized in that, The step of generating the heavy metal recycling decision instruction for metal processing based on the optimization evaluation results includes: Analyze the process parameter configurations in the optimization evaluation results; The process parameter configuration is converted into a sequence of executable instructions for the equipment; The executable instruction sequence of the device is subjected to security verification; By integrating the sequence of executable instructions from the equipment that has passed security verification, the heavy metal recycling decision instructions for the metal processing are obtained.

10. The intelligent decision-making method for heavy metal recovery from wastewater in metal processing as described in claim 1, characterized in that, The triggering of the start / stop control signal for the metal processing equipment based on the heavy metal recycling decision command includes: The process stage sequence in the heavy metal recycling decision instruction is analyzed to obtain the time-series control node set of the metal processing; The device start / stop logic is mapped according to the timing control node set; The start / stop control signal of the metal processing equipment is triggered according to the equipment start / stop logic.

Images