CN-121995750-A - Metal ion efficient leaching system based on multiparameter dynamic cooperation and digital twin

CN121995750ACN 121995750 ACN121995750 ACN 121995750ACN-121995750-A

Abstract

The invention discloses a metal ion efficient leaching system based on multi-parameter dynamic synergy and digital twin, which is characterized in that a digital twin system mapped with a physical leaching system in real time is constructed, prospective instructions are generated by utilizing the prediction deduction capability of the digital twin system, the instructions are converted into real-time adjustment of a dynamic weight distribution mechanism in a multi-parameter cooperative controller through a self-adaptive adjustment algorithm, so that the predictive optimization and closed-loop dynamic adjustment of a leaching process are realized, the unique architecture enables the whole control system to have the self-evolution capability of 'more and more intelligent', a robust control module is arranged in the system, the extreme working conditions such as raw material component fluctuation, sensor data abnormality and actuator failure can be met, stable operation in an industrial environment is ensured, and the core innovation is benefited.

Inventors

ZHOU YUJIE
GUO AIXIN
Yuan Fanxiang
ZHU CHENCHENG
YU YINZHI
JIN JIANJIAO

Assignees

沙洲职业工学院

Dates

Publication Date: 20260508
Application Date: 20260109

Claims (10)

1. The utility model provides a metal ion high-efficient leaching system based on multiparameter dynamic cooperation and digital twin for construct physical-virtual-algorithm closed loop self-evolution framework, realize data-driven closed loop optimization control, characterized by that includes: The online spectrum monitoring module is used for collecting process parameter information of the physical leaching reaction system in the leaching process in real time, wherein the process parameter information comprises metal ion concentration data, pH value and conductivity so as to comprehensively reflect chemical environment changes in the leaching process; The leaching process digital twin module is in communication connection with the online spectrum monitoring module and is used for carrying out simulation deduction on the basis of the acquired data information by the high-fidelity neural network model so as to predict and generate a prediction control instruction containing a future parameter adjustment strategy; The self-adaptive adjustment algorithm module based on deep reinforcement learning is in communication connection with the leaching process digital twin module and the multi-parameter cooperative controller and is used for receiving a prediction control instruction sent by the leaching process digital twin module and converting the prediction control instruction into a current weight adjustment value; The multi-parameter cooperative controller is in communication connection with the self-adaptive adjustment algorithm module and the leaching process digital twin module, and the internal core of the multi-parameter cooperative controller is a cooperative optimization model adopting a dynamic weight distribution mechanism and is used for receiving the current weight adjustment value, updating the weight of the cooperative optimization model in real time according to the current weight adjustment value and outputting a final process parameter control instruction; The intelligent energy consumption management module is in communication connection with the multi-parameter cooperative controller and the leaching process digital twin module and is used for evaluating energy efficiency and feeding back an evaluation result to the leaching process digital twin module so as to iterate the high-fidelity neural network model.
2. The efficient leaching system of metal ions based on multi-parameter dynamic synergy and digital twinning according to claim 1, wherein the dynamic weight distribution mechanism is a weight real-time update mechanism driven by a predictive control instruction of the leaching process digital twinning module, and a mathematical model thereof is expressed as: wherein eta total is the total leaching efficiency, which is the weighted sum of the contribution values of all the technological parameters and represents the whole optimizing effect of the system; w_t (T) is a dynamic weighting factor for the reaction temperature T, which is a function of time T and is not a fixed value, representing the overall optimized weight of the temperature at the current moment; η_t is the contribution value of the reaction temperature T to the overall efficiency; w_t (t) is a dynamic weighting factor for the reaction time t, which is a function of time t, representing the overall optimized weight of the reaction time at the current moment; η_t is the contribution value of the reaction time t to the total efficiency; w_S/L (t) is a dynamic weight factor of the solid-to-liquid ratio S/L, which is a function of time t for representing the weight of the solid-to-liquid ratio optimization as a whole at the current moment; η_S/L is the contribution value of the solid-liquid ratio S/L to the total efficiency; w_ω (t) is a dynamic weighting factor of the stirring speed ω, which is a function of time t, for representing the weight of the stirring speed for the overall optimization at the current moment; η_ω is the contribution value of the stirring speed ω to the total efficiency; In addition, the weight factor w_i (t) is not a fixed value, and is dynamically calculated by an adaptive adjustment algorithm according to a predictive instruction of the digital twin system, so that real-time transfer of an optimized focus based on future trend can be realized.
3. The efficient metal ion leaching system based on multi-parameter dynamic synergy and digital twin according to claim 1 is characterized in that the physical-virtual-algorithm closed loop self-evolution architecture is characterized in that a high-fidelity neural network model is calibrated in real time according to operation data obtained in real time, the training effect of a reinforcement learning algorithm is improved in the accuracy of the digital twin model, the physical leaching reaction system is enabled to operate more stably by an optimized control strategy, higher-quality data are generated, a positive feedback closed loop is formed between the physical leaching reaction system and the digital twin, and the overall performance of the system is continuously improved without manual intervention, wherein the operation data comprise metal ion concentration, temperature, time, solid-liquid ratio, stirring speed, pH value, conductivity and power.
4. The efficient leaching system for metal ions based on multi-parameter dynamic collaboration and digital twin is characterized in that the multi-parameter collaboration controller, the digital twin system and the intelligent energy consumption management system form a closed-loop self-evolution optimization framework, the digital twin system generates a predictive optimization strategy, the multi-parameter collaboration controller performs short-term accurate control, the intelligent energy consumption management system evaluates long-term energy efficiency, an evaluation result is fed back to the digital twin system for model correction and iteration, and the closed-loop self-evolution optimization framework enables the system to autonomously optimize an internal digital twin model and a control strategy through continuous physical-virtual-physical data interaction under the condition that manual intervention is not needed, so that continuous improvement of processing efficiency and energy efficiency is achieved.
5. The efficient leaching system of metal ions based on multi-parameter dynamic synergy and digital twinning according to claim 1, further comprising a robustness control module, wherein the robustness control module acquires real-time sensor data and identifies data anomalies by comparing the real-time sensor data with confidence intervals of predicted values in the leaching process digital twinning module, wherein the real-time sensor data comprises metal ion concentration, temperature, pressure and pH values, and the predicted values are calculated by a high-fidelity neural network model of the leaching process digital twinning module based on the real-time data in a deduction mode, and are predicted values of the system state at present or future time; When the real-time sensor data is not in the confidence interval, the abnormal data point is judged to be ignored temporarily, the short-time maintenance control is carried out according to the last valid data and the digital twin prediction, and meanwhile, the alarm is triggered.
6. The efficient leaching system of metal ions based on multi-parameter dynamic collaboration and digital twin according to claim 5, wherein the robust control module can automatically trigger a safe shutdown procedure when detecting that the process parameters exceed a safe operation window, wherein the safe shutdown procedure comprises cutting off a heating source, stopping stirring and recording fault logs, and the process parameters comprise temperature and pressure.
7. The multi-parameter dynamic synergy and digital twinning-based metal ion efficient leaching system of claim 1, wherein the response time of the adaptive adjustment algorithm module for dynamically adjusting the weight value is less than 1 second.
8. The efficient leaching system for metal ions based on multi-parameter dynamic synergy and digital twin according to claim 1 is characterized in that the construction method of the leaching process digital twin module comprises the steps of training a high-fidelity neural network model by utilizing historical experimental data and a physicochemical mechanism, receiving data of an online spectrum monitoring module and a multi-parameter cooperative controller in real time, keeping the state of the system synchronous with that of a physical leaching reaction system, rapidly simulating the influence of different parameter combinations on final leaching efficiency and total energy consumption in a virtual environment to predict and generate an optimal future parameter adjustment strategy, wherein a prediction time window is 0.5-2 hours in the future, and the accuracy standard of prediction deduction is that the average absolute error of a metal ion concentration predicted value and an actual measured value in the future 1 hour is less than 5%.
9. The efficient leaching system of metal ions based on multi-parameter dynamic synergy and digital twin according to claim 1, wherein the intelligent energy consumption management module adopts a combined optimization strategy of a genetic algorithm and a particle swarm optimization algorithm, and optimizes by taking long-term energy consumption prediction provided by the digital twin system as an objective function.
10. The leaching control method of the metal ion efficient leaching system based on multi-parameter dynamic cooperation and digital twin is characterized by comprising the following steps of: Step S1, an online spectrum monitoring module collects metal ion concentration data in the leaching process in real time; s2, carrying out simulation deduction by the digital twin module in the leaching process according to the obtained metal ion concentration data so as to predict and generate a future parameter adjustment strategy, and sending a prediction control instruction with the future parameter adjustment strategy; Step S3, the self-adaptive adjustment algorithm module calculates a current weight adjustment value of the collaborative optimization model according to a future parameter adjustment strategy in the predictive control instruction; S4, receiving a current weight adjustment value by the multi-parameter cooperative controller, updating a dynamic weight distribution mechanism of a cooperative optimization model in the multi-parameter cooperative controller in real time, calculating to obtain final technological parameters, and outputting a control instruction with the final technological parameters to a physical leaching reaction system; and S5, the intelligent energy consumption management system evaluates the energy efficiency in real time and feeds back an evaluation result to the leaching process digital twin module so as to carry out the next round of model iteration and control.

Description

Metal ion efficient leaching system based on multiparameter dynamic cooperation and digital twin Technical Field The invention relates to the technical field of hydrometallurgy, in particular to a multi-parameter dynamic synergy and digital twin-based metal ion efficient leaching system and method, which are particularly suitable for synchronous, efficient and intelligent recovery of waste lithium ion batteries (including but not limited to LiFePO 4 and NCM, NCA, LCO) and mineral or metallurgical waste materials containing metals such as lithium, iron, cobalt, nickel, manganese and the like. Background The hydrometallurgy field is taken as a traditional industrial field taking a physical and chemical process as a core, the technical development of the hydrometallurgy field is dependent on process improvement, equipment upgrading and reagent optimization for a long time, and a huge field gap exists between the hydrometallurgy field and the digital twin, reinforcement learning and other front information technologies taking a data model and an algorithm as cores. Those skilled in the art generally do not have the knowledge background or motivation to deeply blend these two distinct technical systems, and it is more difficult to predict the subversion effects that may result from their combination. In the prior art, some disclosed leaching methods only focus on optimizing a single process parameter, for example, optimizing only the reaction temperature or the concentration of the leaching agent, while other technical schemes consider a plurality of parameters, generally adopt simple linear combination or step control, fail to establish a dynamic collaborative optimization model among parameters, and further do not have predictive adjustment capability based on prediction, and the control logic of the method basically still belongs to passive response control based on real-time feedback. In summary, the existing leaching technology has the following limitations: 1) Parameter optimization is mostly static, single or linear combination, lacks systematic cooperative consideration, cannot cope with complex nonlinear reaction process, is extremely sensitive to the change of raw material components, and lacks adaptability; 2) The process control is seriously dependent on manual experience, and has large fluctuation of efficiency among batches due to response lag, and lacks effective fault processing and safety guarantee mechanisms when facing extreme working conditions such as sensor faults or abnormal actuators; 3) The energy consumption is high, the treatment cost is high, and a global energy efficiency optimization strategy is lacked; 4) An intelligent control system with prediction and self-learning capabilities is lacking. Disclosure of Invention In order to solve the technical problems, the invention adopts a technical scheme that: The utility model provides a metal ion high-efficiency leaching system based on multiparameter dynamic cooperation and digital twin, which is used for constructing a physical-virtual-algorithm closed-loop self-evolution architecture to realize closed-loop optimization control of data driving, and comprises the following components: The online spectrum monitoring module is used for collecting process parameter information of the physical leaching reaction system in the leaching process in real time, wherein the process parameter information comprises metal ion concentration data, pH value and conductivity so as to comprehensively reflect chemical environment changes in the leaching process; The leaching process digital twin module is in communication connection with the online spectrum monitoring module and is used for carrying out simulation deduction on the basis of the acquired data information by the high-fidelity neural network model so as to predict and generate a prediction control instruction containing a future parameter adjustment strategy; The self-adaptive adjustment algorithm module based on deep reinforcement learning is in communication connection with the leaching process digital twin module and the multi-parameter cooperative controller and is used for receiving a prediction control instruction sent by the leaching process digital twin module and converting the prediction control instruction into a current weight adjustment value; The multi-parameter cooperative controller is in communication connection with the self-adaptive adjustment algorithm module and the leaching process digital twin module, the internal core of the multi-parameter cooperative controller is a cooperative optimization model adopting a dynamic weight distribution mechanism, and is used for receiving the current weight adjustment value, updating the weight of the cooperative optimization model in real time according to the current weight adjustment value (updating the weight in the cooperative optimization model of the internal core of the multi-parameter cooperative controller, namely the weight