Intelligent control system and method for dynamically optimizing copper-silver ore flotation process based on multiple parameters

Abstract

The invention relates to the technical field of mineral processing automation, in particular to an intelligent control system and method for dynamically optimizing a copper-silver ore flotation process based on multiple parameters, and the method comprises the steps that working condition data and technological process parameters of key equipment in the flotation process are obtained; based on the working condition data, constructing and operating an equipment health state digital twin model, and outputting an equipment health state index; a flotation process intelligent control model is built, and a final control instruction is generated; and issuing the final control instruction to key equipment and related execution mechanisms. According to the method, the health state of the execution equipment is quantified in real time through the equipment health state digital twinborn model, when the performance of the equipment is attenuated, the change can be actively sensed and quantified, the control execution deviation caused by the performance degradation of the equipment is compensated from the source, the long-term consistency of the control effect is ensured, and the control efficiency is improved. And the robustness and the environmental adaptability of the whole intelligent control system are obviously enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of mineral processing automation, and particularly relates to a copper-silver ore flotation process intelligent control system and method based on multi-parameter dynamic optimization. BACKGROUND

[0002] Mineral flotation is a complex and variable industrial process, which mainly realizes the maximization of target mineral recovery or the optimization of grade by precisely regulating a series of process parameters, such as reagent dosage, pulp liquid level, aeration amount, stirring intensity, etc. With the development of industrial automation and intelligent technology, it has become a mainstream trend to use a control system based on multi-source data and intelligent algorithms to optimize the flotation process.

[0003] The applicant found in research that in the existing intelligent control scheme, when constructing a control model, the main concern is how to deal with the dynamic changes of external process disturbances such as ore grade, particle size, and pulp concentration. Key equipment such as flotation tanks, mixers, and aeration devices are generally regarded as ideal execution units with constant performance. This processing method ignores the performance degradation of these devices caused by physical factors such as wear, aging, scaling, and clogging during long-term operation. For example, after the wear of the mixer impeller, even if the same speed command is input, the stirring energy and pulp suspension effect will decrease significantly. After the scaling of the aeration element of the aeration device, the same gas supply pressure cannot produce the same bubble dispersion effect.

[0004] This neglect causes the working condition assumptions on which the intelligent control model depends to gradually deviate from the actual state of the physical equipment, that is, the control model and the actual working condition are mismatched. This will cause the optimal control instructions calculated by the intelligent controller to have unexpected deviations in execution when executed by the performance-degraded equipment, which not only weakens the accuracy of the intelligent control system, but also destroys the stability of the closed-loop control, ultimately adversely affecting key economic indicators such as concentrate recovery rate and grade. SUMMARY

[0005] The objective of this application is to provide a method for intelligent control of copper-silver ore flotation process based on multi-parameter dynamic optimization, comprising: acquiring operating condition data of key equipment during operation and process parameters characterizing slurry properties and reaction environment during flotation; constructing and running a digital twin model of equipment health status based on the operating condition data, wherein the digital twin model of equipment health status is used to evaluate in real time the deviation of the actual execution capability of the key equipment from its ideal state due to physical wear, structural aging, or performance degradation, and outputting an equipment health status index that can quantify this deviation; constructing an intelligent control model for the flotation process, wherein the intelligent control model of the flotation process uses the equipment health status index and the process parameters as joint inputs; providing the joint inputs to the intelligent control model of the flotation process for calculation to generate a final control command that can actively compensate for control deviations caused by equipment performance degradation, wherein the final control command is used to adjust one or more operating variables in the flotation process; and issuing the final control command to the key equipment and related actuators to regulate the flotation process in a closed-loop manner, thereby maintaining the stability and optimization of flotation process indicators under conditions of continuous changes in equipment performance.

[0006] By adopting the above technical solution, the health status of the equipment is quantified in real time through a digital twin model of equipment health status, and this health status is used as the input of the intelligent control model of the flotation process. This decouples the control algorithm from the actual physical working conditions of the equipment. When the equipment performance degrades, it can actively sense and quantify this change, and compensate for the control execution deviation caused by the degradation of equipment performance from the root, ensuring the long-term consistency of the control effect and significantly enhancing the robustness and environmental adaptability of the entire intelligent control system.

[0007] Optionally, the construction and operation of the digital twin model of equipment health status includes: when the key equipment is in a baseline health state, collecting its operating condition data under different operating conditions to establish a baseline performance model characterizing its ideal performance; during continuous operation of the equipment, collecting its operating condition data in real time, and using the baseline performance model to calculate the theoretical performance value under the current operating conditions; and calculating the equipment health status index by comparing the theoretical performance value with the actual performance value reflected by the real-time collected operating condition data.

[0008] By adopting the above technical solution, by establishing a benchmark performance model of the execution equipment under ideal conditions and comparing it with the real-time operating conditions of the equipment, standardized equipment health status indicators can be calculated, transforming abstract concepts such as wear and aging into calculable and comparable objective values.

[0009] Optionally, the device health status indicator is a normalized numerical value, and a numerical value variation range of the device health status indicator corresponds to a state evolution process of the device from an initial or ideal health state to a preset performance degradation limit, and is used to represent a real-time attenuation degree of physical performance of the device.

[0010] By adopting the technical solutions described above, Optionally, the key device at least includes one or both of a stirring device or an aeration device in the flotation tank.

[0011] Optionally, when the key device is the stirring device, the generation of the device health status indicator of the stirring device is based on analysis of a dynamic correlation relationship among a motor load parameter, a stirring speed parameter, and a physical property parameter of the ore pulp of the stirring device, so as to evaluate the wear of the stirring impeller or the attenuation of the transmission efficiency.

[0012] Optionally, when the key device is the aeration device, the generation of the device health status indicator of the aeration device is based on analysis of a dynamic correlation relationship between an inlet pressure parameter of a gas supply system and an actual gas flow parameter of the aeration device, so as to evaluate the degree of clogging or fouling of the aeration element.

[0013] Optionally, the operation variable adjusted by the final control instruction at least includes one or more combinations of a speed setting value of the stirring device, an aeration amount setting value of the aeration device, and a dosage amount setting value of the flotation reagent.

[0014] By adopting the technical solutions described above, the wear of the stirring impeller is evaluated by analyzing the relationship between the motor load and the stirring speed, and the clogging of the aeration element is evaluated by analyzing the relationship between the gas supply pressure and the flow, which directly corresponds to the most common physical degradation mode of the device, so that the technical solutions of the present application are no longer an empty theoretical framework, but a specific implementation path with high practicality and industrial application value.

[0015] Optionally, the step of providing the joint input to the flotation process intelligent control model for operation to generate the final control instruction capable of actively compensating for control deviation caused by performance degradation of the device includes a feedforward compensation mechanism, the feedforward compensation mechanism is used to pre-adjust a preliminary control instruction generated by the flotation process intelligent control model in terms of amplitude or form according to the performance degradation degree indicated by the device health status indicator, and the greater the performance degradation degree, the greater the pre-adjustment amplitude.

[0016] By adopting the technical scheme, the device health state index is directly used for pre-adjusting the control instruction through a feedforward compensation mechanism; for example, when a 5% decrease in stirring efficiency is detected, the set value of the stirring speed can be actively increased by a corresponding amplitude before a process deviation is generated. Compared with feedback adjustment after a process index deteriorates, this real-time compensation method has faster response speed, can nip the influence of device performance degradation in the bud, and has more direct and effective control effect.

[0017] Optionally, the method further comprises: pre-setting one or more maintenance warning thresholds associated with the device health state index; continuously monitoring the device health state index, and when the value of the device health state index reaches or is lower than the corresponding maintenance warning threshold, the system automatically generates and issues a predictive maintenance instruction, which contains diagnosis information about the specific device and its performance degradation state.

[0018] By adopting the technical scheme, by setting the maintenance warning threshold, the device health state index can not only be used for process control, but also can automatically generate a predictive maintenance instruction before the device performance degradation affects production or approaches failure, thereby significantly reducing unplanned downtime, optimizing maintenance resource allocation, and thus improving the overall device management level and the operation efficiency of the production line.

[0019] The second object of the present application is to provide an intelligent control system for a copper-silver ore flotation process based on multi-parameter dynamic optimization, comprising: a data acquisition unit configured to acquire real-time operating condition data and process parameters of key equipment in the flotation process; a digital twin modeling unit connected to the data acquisition unit and configured to establish and run a device health state digital twin model based on the operating condition data, and output a device health state index quantifying the degree of performance degradation of the key equipment in real time; an intelligent control decision unit having an input end connected to the data acquisition unit and the digital twin modeling unit, and being configured to receive joint input of the process parameters and the device health state index, and based on the joint input, calculate through an internally integrated intelligent control model of the flotation process to generate a final control instruction that can actively compensate for control deviation caused by device performance degradation; and a control execution unit having an input end connected to an output end of the intelligent control decision unit, and being configured to receive and analyze the final control instruction, and drive the key equipment and related execution mechanisms to perform corresponding physical operations. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 The present application is based on a multi-parameter dynamic optimization copper-silver ore flotation process intelligent control method flowchart.

[0021] Figure 2 The present application is based on a multi-parameter dynamic optimization copper-silver ore flotation process intelligent control system block diagram. DETAILED DESCRIPTION

[0022] In order to make the purposes, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be further described in detail below in combination with specific embodiments. It should be understood that the specific embodiments herein are only used to explain the present application and not to limit the scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0023] In order to overcome the control mismatch problem of the intelligent control model in the prior art due to the neglect of the physical performance degradation of the equipment, the present application proposes an intelligent control method and system that quantifies the equipment health state and integrates it into closed-loop control. By constructing a digital twin model of the equipment health state that is synchronized with the physical equipment in real time, the degree of performance degradation of the equipment caused by factors such as physical wear, structural aging or performance degradation is converted into a calculable and analyzable equipment health state index. The equipment health state index is used as a control variable and is input into the intelligent control model together with the traditional process parameters, thereby generating control instructions that can actively compensate for the performance degradation of the equipment and achieving more robust and adaptive control of the flotation process.

[0024] As shown in Figure 1 The embodiment of the present application discloses an intelligent control method for the flotation process of copper-silver ore based on multi-parameter dynamic optimization. The method is applied to a typical copper-silver ore flotation production line to stabilize and improve the recovery rate or concentrate grade of copper and silver. The method includes the following steps.

[0025] S01: Obtain the working condition data of the key equipment in the flotation process and the process parameters representing the properties of the ore pulp and the reaction environment.

[0026] It can be understood that in the present embodiment, the key equipment of interest is mainly the stirring device and the aeration device in the flotation tank, because their performance state directly affects the uniformity of the suspension of the ore pulp, the aeration dispersion effect and the contact efficiency of the reagent and the ore particles, and they are the key execution mechanism of the flotation process.

[0027] Specifically, for agitation devices, the required operating data includes motor load parameters and agitation speed parameters. For example, a high-precision intelligent power meter or current transformer installed on the power supply circuit of the agitator's drive motor can collect the motor's active power, reactive power, current, and voltage in real time at a sampling frequency of no less than 5Hz. These parameters comprehensively reflect the resistance overcome by the motor when driving the agitator impeller to rotate, and are important data characterizing agitation power consumption. Alternatively, a rotary encoder or Hall sensor installed on the motor shaft or gearbox output shaft can obtain the actual rotation speed of the agitator in real time. This parameter is used to confirm whether control commands are executed accurately and serves as an important input for calculating agitation efficiency.

[0028] For the aeration device, the operating data to be collected includes the inlet pressure parameters of the gas supply system and the actual gas flow rate parameters. For example, a pressure transmitter is installed on the main outlet pipeline of the blower or on the branch pipeline entering each flotation cell to monitor the gas supply pressure in real time. In the aeration pipeline entering the flotation cell, a mass flow meter or vortex flow meter is installed to accurately measure the actual volume or mass of gas entering the slurry per unit time.

[0029] Process parameters characterize the physicochemical properties of the slurry and the macroscopic environment of the flotation reaction, and are the main inputs to traditional intelligent flotation control models. Process parameters include slurry property parameters, reaction environment parameters, and operating variable parameters.

[0030] Specifically, slurry property parameters can be obtained through online analytical instruments. For example, ultrasonic concentration meters or nuclear density meters can be used to measure slurry concentration (or density); online particle size analyzers can be used to monitor the particle size distribution of the feed ore; and X-ray fluorescence analyzers can be used to detect the grades of elements such as copper and silver in the raw ore, concentrate, and tailings online. Reaction environment parameters can be obtained through sensors installed in the flotation cell. For example, pH meters can be used to monitor the acidity or alkalinity of the slurry; redox potential electrodes can be used to monitor the electrochemical environment of the slurry; and level gauges or pressure sensors can be used to measure the slurry level in the flotation cell. Operational variable parameters sent from the current control system to the actuators can also be collected, such as the stroke frequency or rotation speed of the reagent dosing pump corresponding to the reagent dosage, the output frequency of the frequency converter corresponding to the stirring speed, and the opening degree of the aeration valve regulator.

[0031] Understandably, all collected data can be given a precise timestamp and transmitted in real time to the subsequent data processing and modeling unit via industrial Ethernet or fieldbus.

[0032] S02: Based on operating data, construct and run a digital twin model of equipment health status. The digital twin model of equipment health status is used to assess in real time the degree of deviation of the actual performance capability of key equipment from its ideal state due to physical wear, structural aging or performance degradation, and output equipment health status indicators that can quantify this degree of deviation.

[0033] Specifically, constructing and running a digital twin model of equipment health status includes: collecting operating data of key equipment under different operating conditions when the equipment is in a baseline health state to establish a baseline performance model that characterizes its ideal performance; collecting operating data in real time during continuous equipment operation and using the baseline performance model to calculate the theoretical performance value under the current operating conditions; and calculating the equipment health status index by comparing the theoretical performance value with the actual performance value reflected by the real-time collected operating data.

[0034] Understandably, benchmark performance models are typically established when the equipment is installed, commissioned, and in a brand-new, ideal, and healthy state, or after a complete overhaul, such as when a brand-new impeller or aeration element has been replaced. The operation method involves manually changing the equipment's operating commands in a stepwise or ramp manner within a relatively wide operating range and recording the corresponding steady-state operating data.

[0035] Specifically, taking a mixing device as an example, under the premise that the slurry properties are relatively stable, for example, using water or tailings slurry of a specific concentration for testing, the mixing speed setting is gradually increased in multiple steps from the lowest speed to the highest speed. Each step can be defined as an increase of 5% of the rated speed. At each speed step, after the motor load parameters stabilize, the current mixing speed and active power data are recorded. By performing statistical regression analysis on the large amount of collected data, for example using polynomial fitting, support vector regression, etc., a mathematical model can be established. This mathematical model describes the functional relationship between the active power of the mixing device and the mixing speed under ideal healthy conditions. For example, ideal active power = f(mixing speed, slurry property parameters). This function f is the benchmark performance model of the mixing device, defining how much energy a healthy mixing device should consume under specific speed and slurry conditions.

[0036] Taking the aeration device as an example, under the condition that the liquid level in the flotation cell is stable, the opening of the aeration control valve is gradually adjusted from the minimum to the maximum, and a series of stable inlet pressure and actual gas flow data pairs of the gas supply system are recorded. Through regression analysis, a function describing the relationship between the two is established, such as ideal gas flow = g(gas supply pressure, slurry liquid level). This function g is the benchmark performance model of the aeration device, which defines how much gas a smooth aeration element should be able to pass through under a specific gas supply pressure and immersion depth.

[0037] It can be understood that after the device enters the long-term continuous operation stage, the device health status digital twin model starts to work in real time.

[0038] The device health status digital twin model performs the following steps in each calculation cycle: a. Obtain the current real-time operating data of the device from the data acquisition unit, for example, for the stirring device, obtain its real-time rotating speed and real-time active power, etc. real-time operating parameters; for the aerator, obtain its real-time pressure and real-time flow, etc. real-time operating parameters.

[0039] b. Substitute the above-mentioned real-time operating parameters and related process parameters into the previously established benchmark performance model, wherein the process parameters can be ore pulp density, liquid level, etc.

[0040] c. Calculate the theoretical performance value that the device should have under the real-time operating conditions if it is in an ideal healthy state.

[0041] d. Compare the theoretical performance value with the actually measured performance value, and calculate the standardized device health status index, which is designed to change between 0 and 1 (or 0% to 100%), and the numerical value change range corresponds to the state evolution process of the device from the initial or ideal healthy state to the pre-set performance degradation limit, used to represent the real-time attenuation degree of the physical performance of the device, wherein 1 (or 100%) represents the ideal healthy state, and the smaller the value, the more serious the performance attenuation.

[0042] It can be understood that for the stirring device, as the impeller wears out, its ability to push the ore pulp to do work decreases, so at the same rotating speed, the fluid resistance overcome by the motor will decrease, resulting in actual active power consumption lower than the theoretical active power. Therefore, its health status index can be defined as: stirring health index = actual active power consumption / theoretical active power; for example, the calculated stirring health index is 0.92, which means that the current stirring efficiency of the stirring device is only about 92% of the ideal state, with 8% performance degradation.

[0043] For the aerator, as the aeration element is scaled or blocked, the resistance of the gas passing increases, resulting in actual gas flow passing at the same inlet pressure lower than the theoretical gas flow; therefore, its health status index can be defined as: aeration health index = actual gas flow passing / theoretical gas flow; for example, the calculated aeration health index is 0.85, which means that the aeration capacity of the aerator is only 85% of the ideal state, with a significant degree of blockage.

[0044] Through the above steps, the abstract device wear or performance degradation concept is successfully converted into a specific value that can be used for automatic control, i.e. the device health status index.

[0045] S03: Constructing a flotation process intelligent control model, which takes the equipment health state indicators and the process parameters as joint inputs.

[0046] Unlike the prior art, the intelligent control model of the present application takes the equipment health state indicators as a brand-new important input dimension.

[0047] Specifically, the flotation process intelligent control model can be implemented by using various advanced control algorithms, such as mechanism and data hybrid modeling, expert system, fuzzy logic control, or agent based on reinforcement learning. In the present embodiment, a neural network based predictive control model can be used.

[0048] The input layer of the flotation process intelligent control model receives joint inputs, which at least include process parameters and equipment health state indicators, wherein the process parameters include ore grade, particle size, pulp concentration, pH value, etc., and the equipment health state indicators include the above-mentioned agitation health indicators and aeration health indicators.

[0049] It can be understood that by collecting a large amount of historical production data containing the above-mentioned joint inputs and corresponding flotation results (such as concentrate grade and recovery rate), the neural network model is trained offline, and the training goal is to let the neural network model learn what operation should be taken under different combinations of ore characteristics, reaction environment and equipment health state to achieve the optimal flotation effect, such as adjusting the stirring speed, aeration amount, reagent dosage, etc.; through learning, the model will inherently understand that when the agitation health indicator is low, a higher stirring speed setting value is needed than normal to achieve the same pulp suspension effect.

[0050] S04: Providing the joint inputs to the flotation process intelligent control model for operation to generate final control instructions that can actively compensate for the control deviation caused by the equipment performance degradation, and the final control instructions are used to adjust one or more operation variables in the flotation process.

[0051] Specifically, in each control cycle, the intelligent control decision unit first inputs the real-time acquired process parameters into the trained flotation process intelligent control model; at this time, it can be temporarily assumed that the equipment is healthy, i.e., the health state indicator is 1.0, and the flotation process intelligent control model will calculate a set of preliminary control instructions under ideal conditions; for example, the flotation process intelligent control model judges that the current ore grade is rising, and the agitation needs to be strengthened, and the preliminary instruction given is to set the stirring speed to 400 rpm.

[0052] Next, a feedforward compensation mechanism is introduced, which is to pre-adjust the preliminary control instructions generated by the intelligent control model of the flotation process in amplitude or form according to the performance degradation degree indicated by the equipment health state indicator, wherein the greater the performance degradation degree, the greater the pre-adjustment amplitude.

[0053] Then, a compensation factor is calculated, wherein the compensation factor is a function of the health state indicator, usually its inverse or calculated based on other rules, the more serious the performance degradation, that is, the lower the health indicator, the greater the compensation factor; for example, the agitation compensation factor = 1 / agitation health indicator, the aeration compensation factor = 1 / aeration health indicator.

[0054] Finally, the compensated control instruction is generated: multiply the preliminary control instruction by the compensation factor, or calculate or combine in other forms to obtain the compensated control instruction; for example, the compensated agitation speed instruction = preliminary agitation speed instruction * agitation compensation factor, the compensated aeration amount instruction = preliminary aeration amount instruction * aeration compensation factor.

[0055] For example, assuming that the above calculation shows that the current agitation health indicator is 0.80, that is, the efficiency has decreased by 20%, then the agitation compensation factor is 1 / 0.80 = 1.25, and the system will multiply the preliminary instruction of 400 rpm by 1.25 to obtain the compensated instruction of 500 rpm.

[0056] This process is feedforward, that is, the amplitude of the control instruction is actively amplified at the moment when the equipment execution capability is perceived to decrease, rather than waiting for the lagging results such as uneven suspension of ore pulp and decrease of recovery rate to appear for feedback adjustment, so as to execute the instruction with 25% amplification intensity on the equipment with 80% performance, and finally achieve the same physical effect as the healthy equipment executing the original instruction.

[0057] The compensated control instruction, combined with other control instructions generated directly by the intelligent control model or unrelated to the health of the equipment, such as the dosage of reagents, constitutes the final control instruction, which is the optimal operation strategy under the current working condition of the equipment health condition. The final control instruction is used to adjust at least one or more combinations of the following operation variables: the speed setting value of the agitation device, the aeration amount setting value of the aeration device, and the dosage setting value of the flotation reagent.

[0058] S05: The final control instruction is issued to the key equipment and related execution mechanism to regulate the flotation process in a closed loop, so as to maintain the stability and optimization of the flotation process indicators under the condition of continuous change of equipment performance.

[0059] The intelligent control decision unit sends the set of final control instructions to the control execution unit through a communication interface. For example, the final control instructions are: the stirring speed is set to 500 rpm, the aeration valve opening is set to 65%, and the collector A addition rate is 150 ml / min.

[0060] The control execution unit is usually a control module of a programmable logic controller or a distributed control system. The control execution unit parses these final control instructions and converts them into corresponding physical driving signals, such as sending an analog signal or a digital set value to the frequency converter that controls the stirring motor, so that the motor operates at a target speed of 500 rpm; sending instructions to the electrical positioner that controls the aeration pipeline valve to adjust the valve opening to 65%; and sending instructions to the controller that controls the reagent addition pump to set its working frequency to achieve a dosing rate of 150 ml / min.

[0061] Through the above five-step cycle, the system forms an intelligent closed-loop control system that can dynamically perceive and actively adapt to changes in device performance. Even if the stirring impeller gradually wears out and the aeration element slowly fouls during weeks of operation, the control system can always maintain precise and stable control over the flotation process by continuously adjusting the compensation factor, thereby maintaining efficient operation of the copper-silver ore flotation process for a long time.

[0062] It can be understood that the above method can also include a predictive maintenance step, which specifically includes: pre-setting one or more maintenance warning thresholds associated with the device health status indicators, continuously monitoring the device health status indicators, and when the value of the device health status indicators reaches or is below the corresponding maintenance warning threshold, the system automatically generates and issues a predictive maintenance instruction, which contains diagnostic information about the specific device and its performance degradation state.

[0063] Specifically, according to production experience, equipment maintenance manual and economic considerations, multiple levels of thresholds are set for the health status indicators of key equipment. For example: health status (90%-100%): the equipment is in good condition and does not require intervention; attention status (80%-90%): the equipment has slight performance degradation, and the system marks attention but does not issue instructions for the time being; warning threshold (70%): when the health indicator is below 70%, it indicates that the performance degradation is relatively obvious and may begin to affect the control limit and energy economy, so maintenance is needed; forced maintenance threshold (60%): when the health indicator is below 60%, it indicates that the device performance has deteriorated seriously, the control compensation may reach the upper limit, or there is a risk of operation, and maintenance must be arranged in the near future.

[0064] It can be understood that the system will continuously record the time series data of each device health status indicator, and by applying time series analysis models such as autoregressive integrated moving average models or simple linear extrapolation to these historical data, the declining trend of the health status indicator in the future can be predicted.

[0065] When the system predicts that at a certain future time point, for example, it is expected that 15 days later, the stirring health indicator of a stirring device will drop below the early warning threshold of 70%, it will automatically generate a predictive maintenance instruction.

[0066] The predictive maintenance instruction will be sent to the device maintenance department through the factory's manufacturing execution system or device management system. The content of the predictive maintenance instruction is clear and explicit, for example: "Warning: flotation plant series 2 No. 3 flotation tank stirring device, device ID: F-203, current health indicator is 72%, predicted to drop below 70% early warning line within 15 days. Diagnosis information: stirring impeller wear causes efficiency decline. Suggestion: please arrange for replacement of impeller operation in the next planned shutdown maintenance window, and prepare spare parts in advance, spare parts number: JB-YL-007." This predictive maintenance mechanism greatly reduces unplanned downtime due to sudden device failure, optimizes spare parts inventory, and improves overall production line efficiency and device comprehensive efficiency.

[0067] As shown in Figure 2 The embodiment of the application discloses an intelligent control system for copper-silver ore flotation process based on multi-parameter dynamic optimization, which comprises: a data acquisition unit configured to acquire real-time working condition data and process parameters of key equipment in the flotation process; a digital twin modeling unit connected with the data acquisition unit, configured to establish and run a device health status digital twin model based on the working condition data, and output a device health status indicator quantifying the degree of performance degradation of the key equipment in real time; an intelligent control decision unit, the input end of which is connected with the data acquisition unit and the digital twin modeling unit, configured to receive joint input of the process parameters and the device health status indicator, and based on the joint input, calculate through an internally integrated intelligent control model of the flotation process to generate a final control instruction that can actively compensate for control deviation caused by device performance degradation; and a control execution unit, the input end of which is connected with the output end of the intelligent control decision unit, configured to receive and analyze the final control instruction, and drive the key equipment and related execution mechanisms to perform corresponding physical operations.

[0068] Specifically, the data acquisition unit is composed of various hardware devices deployed in the production site; the sensor layer includes current transformers, rotary encoders, pressure transmitters, mass flow meters, pH meters, level meters, online grade analyzers, etc., which are responsible for converting physical or chemical quantities into electrical signals. The data acquisition and transmission layer is composed of remote I / O modules, programmable logic controllers or field control stations of distributed control systems, which are responsible for collecting raw electrical signals from sensors, performing preliminary signal conditioning such as filtering, amplification, analog-to-digital conversion, and packaging structured data according to a unified communication protocol, sending it to the upper unit through an industrial Ethernet composed of industrial switches and optical fibers. The data acquisition unit scans all measurement points uninterruptedly with a cycle of milliseconds to seconds, forming a real-time data stream covering the entire flotation process.

[0069] The digital twin modeling unit is usually deployed in the form of software on an edge computing server or an industrial PC; it can be a dedicated software service or a microservice integrated in the platform, containing a real-time database for caching short-time historical data received from the data acquisition unit; it also includes a model library that internally stores benchmark performance models for all key devices, such as ideal active power = f(...) and ideal gas flow = g(...), etc., which are fixed after initial calibration, and a computing engine responsible for executing the calculation logic of the health status indicators.

[0070] Specifically, the digital twin modeling unit subscribes to the relevant operating condition data published by the data acquisition unit; every other preset period, such as 1 minute, the computing engine triggers a calculation task; for each monitored device, the computing engine extracts its latest operating condition data from the real-time database; calls the corresponding benchmark performance model in the model library to calculate the theoretical performance value; calculates the standardized device health status indicators according to the preset formula, such as actual value / theoretical value; publishes the calculated health status indicators along with the device ID and timestamp to the system's message bus or directly sends them to the intelligent control decision unit.

[0071] The intelligent control decision unit is usually deployed on a server or cloud platform with stronger performance to obtain powerful computing power, such as hardware configuration with high-performance multi-core CPU and large-capacity memory, and some scenarios can be configured with GPU to accelerate the operation of neural network models; the flotation process intelligent control model includes the above-mentioned neural network predictive control model, reinforcement learning agent or other advanced control algorithms; the feedforward compensation module is used to calculate the compensation factor according to the input health status indicators and adjust the preliminary instructions; the predictive maintenance module integrates time series analysis and prediction algorithms, and manages the maintenance warning threshold rule library; the decision fusion and instruction generation engine is responsible for coordinating the outputs of each module to form the final control instructions.

[0072] Specifically, the intelligent control decision unit receives real-time process parameters from the data acquisition unit and equipment health status indicators from the digital twin modeling unit; these joint inputs are fed into the flotation process intelligent control model to calculate preliminary control instructions; at the same time, the equipment health status indicators are fed into the feedforward compensation module to calculate compensation factors; the decision engine applies the compensation factors to adjust the preliminary instructions to compensated instructions; all instructions are fused to generate a final control instruction set; the final control instruction is output to the control execution unit; in parallel, the predictive maintenance module continuously analyzes the time series of health indicators and independently generates and issues predictive maintenance instructions when conditions are met. The control execution unit is the bridge connecting the information world and the physical world, and is usually the DCS or PLC system itself in the field. The controller is the controller of the DCS or an independent PLC that receives digital instructions from the intelligent control decision unit. The actuator includes a frequency converter, a servo driver, an intelligent valve positioner, a metering pump controller, etc. connected with the controller. The physical equipment is the final driven stirring device motor, air valve, reagent pump, etc.

[0073] Specifically, the controller receives the final control instructions issued by the upper layer, such as speed setpoint, valve opening, etc.; the control logic inside the controller converts these high-level setpoints into low-level output signals; these output signals are sent to the corresponding actuators; the actuators drive the physical equipment to perform precise physical operations according to the instructions, thereby closing the loop of intelligent decision-making to the actual production process.

[0074] In summary, the present application successfully includes the equipment physical performance degradation, a traditional control blind area, into the intelligent control loop by introducing the equipment health status digital twin model. Not only does it provide a method for real-time compensation of equipment performance degradation, improving control accuracy and robustness, but also optimizes equipment management mode through predictive maintenance function. The method and system architecture are clear, the logic is rigorous, and the functions of each part are clear, providing a complete, effective and highly industrial application value technical solution for solving the control problems caused by equipment aging in complex industrial processes.

[0075] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional modules is exemplified, and in actual application, the above functions can be completed by different functional modules according to needs, i.e. the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0076] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other manners. For example, the described device embodiments are merely schematic. Taking the division of the modules or units as an example, the division can not necessarily be in accordance with an actual physical division, and there can be one or more intermediate modules or units, and one or more specific functions of modules or units can be combined or integrated into another system, and the original function of the module or unit can be performed or implemented. In addition, the display or discussion about a directly coupled or indirect coupled or communication connection can be implemented through another interface, device or unit, and the electrically, mechanically or other forms of direct or indirect coupling or communication connection can be implemented using other forms.

[0077] The units described as separate components may or can not be physically separate, and the components shown as units may or can not be physical units, i.e., they can be located in one place or distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0078] In addition, the functional units in each embodiment of the present application can be integrated into one processing unit, or each unit can be physically present separately, or two or more units can be integrated into one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0079] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part of the prior art that contributes to the technical solutions or all or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory, a random access memory, a magnetic disk or an optical disk, and various media that can store program codes.

[0080] The above are preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any feature disclosed in the specification (including the abstract and drawings) can be replaced by other equivalent or similar features unless specifically stated. That is, each feature is only an example of a series of equivalent or similar features unless specifically stated.

Claims

1. A multi-parameter based dynamic optimization intelligent control method for copper-silver ore flotation process, characterized in that, The method comprises: acquiring the working condition data of key equipment in a flotation process and process parameters representing the properties of ore pulp and reaction environment; based on the working condition data, constructing and running a digital twin model of equipment health state, which is used to evaluate the deviation of actual execution capacity of the key equipment from its ideal state due to physical wear, structural aging or performance degradation, and output an equipment health state indicator quantifying the deviation; constructing an intelligent control model of the flotation process, which takes the equipment health state indicator and the process parameters as joint inputs; providing the joint inputs to the intelligent control model of the flotation process for operation to generate a final control instruction that can actively compensate for control deviation caused by equipment performance degradation, which is used to adjust one or more operation variables in the flotation process; issuing the final control instruction to the key equipment and related execution mechanism to regulate the flotation process in a closed loop, thereby maintaining the stability and optimization of flotation process indicators under the condition of continuous change of equipment performance.

2. The method of claim 1, wherein, The construction and running of the digital twin model of equipment health state comprises: when the key equipment is in a baseline health state, collecting its working condition data under different operating conditions to establish a baseline performance model representing its ideal performance; during the continuous operation of the equipment, collecting its working condition data in real time and calculating the theoretical performance value under the current operating condition by using the baseline performance model; by comparing the theoretical performance value with the actual performance value reflected by the real-time collected working condition data, the equipment health state indicator is calculated.

3. The method of claim 2, wherein, The equipment health state indicator is a standardized numerical value, and its numerical value change range corresponds to the state evolution process of the equipment from the initial or ideal health state to the preset performance degradation limit, which is used to represent the real-time degradation degree of the physical performance of the equipment.

4. The method of claim 1, wherein, The key equipment at least includes one or both of the stirring device or the aeration device in the flotation tank.

5. The method of claim 4, wherein, When the key equipment is the stirring device, the generation of its equipment health state indicator is based on the analysis of the dynamic correlation between the motor load parameter, the stirring speed parameter of the stirring device and the physical property parameter of the ore pulp, to evaluate the wear or transmission efficiency degradation of the stirring impeller.

6. The method of claim 4, wherein, When the key equipment is the aeration device, the generation of its equipment health state indicator is based on the analysis of the dynamic correlation between the gas supply system inlet pressure parameter and the actual gas flow parameter of the aeration device, to evaluate the degree of blockage or fouling of the aeration element.

7. The method of claim 1, wherein, The final control instruction is used to adjust at least one or more combinations of the following operation variables: the speed setting value of the stirring device, the aeration amount setting value of the aeration device, and the dosage amount setting value of the flotation reagent.

8. The method of claim 7, wherein, The step of providing the joint input to the flotation process intelligent control model for calculation to generate the final control instruction capable of actively compensating for the control deviation caused by equipment performance degradation contains a feedforward compensation mechanism, which is a preliminary adjustment in amplitude or form of the preliminary control instruction generated by the flotation process intelligent control model according to the performance degradation degree indicated by the equipment health state indicator, wherein the greater the performance degradation degree, the greater the amplitude of the preliminary adjustment.

9. The method of claim 1, wherein, Also comprising: pre-setting one or more maintenance warning thresholds associated with the equipment health state indicator; continuously monitoring the equipment health state indicator, and when the value of the equipment health state indicator reaches or is lower than the corresponding maintenance warning threshold, the system automatically generates and issues a predictive maintenance instruction, which contains diagnostic information of the specific equipment and its performance degradation state.

10. A multi-parameter based dynamic optimization intelligent control system for copper-silver ore flotation process, characterized in that, Comprising: a data acquisition unit configured to acquire real-time working condition data and process parameters of key equipment in the flotation process; a digital twin modeling unit connected with the data acquisition unit, configured to establish and run an equipment health state digital twin model based on the working condition data, and output an equipment health state indicator quantifying the performance degradation degree of the key equipment in real time; an intelligent control decision unit, the input end of which is connected with the data acquisition unit and the digital twin modeling unit, configured to receive the joint input of the process parameters and the equipment health state indicator, and based on the joint input, calculate through an internally integrated flotation process intelligent control model to generate a final control instruction capable of actively compensating for the control deviation caused by equipment performance degradation; a control execution unit, the input end of which is connected with the output end of the intelligent control decision unit, configured to receive and analyze the final control instruction, and drive the key equipment and related execution mechanisms to perform corresponding physical operations.

Images