A gold ore refining intelligent control system and method

Abstract

The present invention relates to the field of gold ore refining, and in particular to an intelligent control system and method for gold ore refining. The content includes: processing refining data through adaptive filtering and dynamic average processing algorithm to obtain smoothed refining data; based on the smoothed refining data, using a gold ore refining feature extraction algorithm to process and obtain gold ore refining features; based on the gold ore refining features, calculating the gold recovery rate through a gold recovery rate dynamic calculation algorithm; based on the gold recovery rate, using a gold ore refining process optimization algorithm to optimize process parameters, obtaining updated values ​​of process parameters, and adjusting actual process parameters according to the updated values ​​of process parameters. The problem of failing to accurately consider the dependency between time and features, ignoring the long-term cumulative effect of features, failing to adjust parameters in real time in a complex smelting environment, and ignoring that the adjustment of process parameters must follow a certain physical limit range is solved in the gold ore refining process.

Description

Technical Field

[0001] The present invention relates to the field of gold ore refining, and in particular to an intelligent control system and method for gold ore refining. Background Art

[0002] Gold ore refining is a key link in metal mineral processing, especially for the extraction of precious metals (such as gold and silver). It has important economic significance. The gold ore refining process mainly includes ore crushing, grinding, leaching, enrichment, purification, refining and other stages. Each stage requires strict control of process parameters and operating conditions to ensure efficient metal recovery rate and high purity of the final product.

[0003] In the traditional gold ore refining process, the operation of each link needs to rely on manual or simple automated equipment for control. However, with the increasing shortage of mineral resources and the demand for efficient use of resources, the control accuracy and efficiency in the refining process are particularly important. Traditional gold ore refining control methods have certain limitations in improving production efficiency and control accuracy. The use of intelligent control technology to improve the efficiency of the gold ore refining process, reduce energy consumption, and improve product quality has become an important development direction in the mining industry. A gold ore refining intelligent control method has emerged. By introducing intelligent control, the automation, refinement and optimization of the refining process can be achieved, the metal recovery rate can be significantly improved, energy consumption can be reduced, and environmental pollution can be reduced, which will promote the gold ore refining industry to develop in the direction of efficiency, greenness and intelligence. Summary of the invention

[0004] The present invention provides an intelligent control system and method for gold ore refining, so as to solve the problems that, in the process of gold ore refining, the dependency between time and characteristics is not accurately considered, thus affecting the effective control of the refining process; the long-term cumulative effect of characteristics is ignored, and the gold recovery rate in the process of gold ore refining cannot be accurately predicted and optimized; the parameters cannot be adjusted in real time in a complex smelting environment, and the adjustment of process parameters must comply with a certain physical limit range, which causes the process parameters to exceed the limit of the equipment or process reaction, thus bringing safety hazards.

[0005] The present invention provides an intelligent control system and method for gold ore refining, which specifically includes the following technical solutions:

[0006] A gold ore refining intelligent control method comprises the following steps:

[0007] S1: collect refined data in real time, and process the refined data through adaptive filtering and dynamic averaging processing algorithm to obtain refined data after smoothing;

[0008] S2: Based on the refined data after smoothing, the gold ore refining feature extraction algorithm is used to obtain the gold ore refining feature;

[0009] S3: Calculate the gold recovery rate through the gold recovery rate dynamic calculation algorithm based on the gold ore refining characteristics;

[0010] S4: Based on the gold recovery rate, the process parameters are optimized using the gold ore refining process optimization algorithm to obtain updated values ​​of the process parameters, and the actual process parameters are adjusted according to the updated values ​​of the process parameters.

[0011] Preferably, the S1 specifically includes:

[0012] In the process of implementing the adaptive filtering and dynamic averaging processing algorithm, the average value of the refined data at each time point is obtained based on the refined data at the current time point and the historical refined data through the dynamic averaging method; the smoothing degree of the current refined data is dynamically adjusted according to the difference between the refined data at each time point and the average value of the refined data through the adaptive filtering technology, and the influence of the historical refined data on the current refined data is controlled by the attenuation factor; at the same time, the smoothing degree of the refined data at each time point is adjusted by introducing a control parameter of the smoothing degree; the calculation formula of the adaptive filtering and dynamic averaging processing algorithm is:

[0013] ;

[0014] in, Indicated in After time point smoothing Refined data; Indicated in The time point Refined data; Indicated in The time point Refined data and running average The absolute difference between A control parameter indicating the degree of smoothing; Indicates The attenuation factor related to the refined data is used to control the effect of historical refined data on The time point The impact of refined data; Indicated in The time point The dynamic average of each refined data is calculated based on the historical refined data.

[0015] Preferably, the S2 specifically includes:

[0016] The gold ore refining feature extraction algorithm performs weighted processing on each smoothed refined data by using a weighting coefficient, and simulates the periodicity of the smoothed refined data by a cosine function; at the same time, frequency parameters corresponding to different refining features are introduced to control the period of the cosine function.

[0017] Preferably, the S3 specifically includes:

[0018] The dynamic calculation algorithm of gold recovery rate adopts nonlinear weighted combination method to gold ore refining characteristics, combining the immediate impact and cumulative effect of refining characteristics to calculate the gold recovery rate; the calculation formula of gold recovery rate is:

[0019] ;

[0020] in, Indicated in Gold recovery at the time point; Indicates the number of refined features; Indicates The weight coefficient of the refined features; Indicated in The time point A refined feature; Indicates The attenuation coefficient of the refined features; is the cumulative effect term, indicating that The integral of a refined feature from the previous time point to the current time point reflects the cumulative effect of the refined features; is the variable of integration; Indicated in The time point A refined feature.

[0021] Preferably, the S4 specifically includes:

[0022] The gold ore refining process optimization algorithm constructs an objective function and uses an adaptive learning rate mechanism to optimize process parameters to achieve optimal control of the gold ore refining process.

[0023] Preferably, the S4 specifically includes:

[0024] The objective function is constructed by calculating the difference between the gold recovery rate and the target gold recovery rate and the difference between the production cost and the target production cost, and combining the exponential decay term and the smooth constraint term; the objective function formula is:

[0025] ;

[0026] in, represents the objective function; Indicated in The difference between the gold recovery rate at a certain point in time and the target gold recovery rate is calculated using the formula: , represents the target gold recovery rate; Indicated in The difference between the production cost at a point in time and the target production cost is calculated as , Indicated in Production costs at a point in time, represents the target production cost; and The weight coefficients respectively represent the relative importance of adjusting the gold recovery rate difference and the production cost difference in the objective function; represents the exponential decay term; represents the constant that controls the attenuation strength; represents the smoothness constraint; Indicated in The time point The value of a process parameter; Indicated in The time point The value of a process parameter; Indicates the number of process parameters; Represents the process parameter adjustment coefficient.

[0027] Preferably, the S4 specifically includes:

[0028] The gold ore refining process optimization algorithm uses an adaptive learning rate mechanism to update the process parameters and obtain the updated values ​​of the process parameters; the calculation formula is:

[0029] ;

[0030] in, Indicated in The time point Updated values ​​of process parameters; Indicated in The objective function at the time point is The gradient of the process parameters; Indicated in The learning rate at this time step.

[0031] Preferably, the S4 specifically includes:

[0032] Constraints are set so that the updated values ​​of the process parameters follow physical limits that are governed not only by the maximum capacity of the equipment but also by the chemical and physical characteristics of the refining reactions.

[0033] An intelligent control system for gold ore refining, comprising the following parts:

[0034] Data acquisition module, data smoothing module, feature extraction module, gold recovery rate calculation module, process parameter optimization module, process parameter control module;

[0035] Data acquisition module: collects refined data in real time and outputs the refined data to the data smoothing processing module;

[0036] Data smoothing processing module: Adopts adaptive filtering and dynamic average processing algorithm to process refined data, obtains refined data after smoothing, and outputs the refined data after smoothing to the feature extraction module;

[0037] Feature extraction module: Based on the refined data after smoothing, the gold ore refining feature extraction algorithm is used to obtain the gold ore refining feature, and the gold ore refining feature is output to the gold recovery rate calculation module;

[0038] Gold recovery rate calculation module: calculates the gold ore refining characteristics through the gold recovery rate dynamic calculation algorithm to obtain the gold recovery rate, and outputs the gold recovery rate to the process parameter optimization module;

[0039] Process parameter optimization module: Based on the gold recovery rate of the gold recovery rate calculation module, the process parameters in the refining process are optimized using the gold ore refining process optimization algorithm to obtain updated values ​​of the process parameters, and the updated values ​​of the process parameters are output to the process parameter control module;

[0040] Process parameter control module: adjusts the actual process parameters according to the process parameter update values ​​of the process parameter optimization module.

[0041] The beneficial effects of the technical solution of the present invention are:

[0042] 1. By using adaptive filtering and dynamic averaging processing algorithms to smooth the refined data, the noise and outliers in the refined data are eliminated, which can ensure that the refined data is accurate and stable, and provide a reliable data basis for subsequent analysis.

[0043] 2. Based on the refined data after smoothing, the gold ore refining feature extraction algorithm is adopted, combined with weighted coefficients and periodic analysis, to accurately extract the key features of the gold ore refining process. By modeling the phase adjustment and periodic changes of different features, the dynamic relationship between different features can be revealed, and the noise interference that may exist in the time series can be eliminated, thereby enhancing the accuracy and stability of feature extraction.

[0044] 3. Use the dynamic calculation algorithm of gold recovery rate to perform nonlinear weighted combination of gold ore refining characteristics, and consider the immediate impact and cumulative effect of the characteristics to accurately calculate the gold recovery rate; compared with the traditional instant gold recovery rate calculation, the present invention comprehensively considers the long-term impact of various characteristics on the gold recovery rate, and introduces the cumulative effect term to dynamically reflect the impact of each stage in the refining process on the gold recovery rate, effectively improving the accuracy of the gold recovery rate.

[0045] 4. Based on the gold recovery rate, the gold ore refining process optimization algorithm is used to optimize the key process parameters in the refining process to achieve the dual goals of maximizing gold recovery rate and minimizing production costs; through the adaptive learning rate mechanism, the learning rate converges quickly in the initial stage of optimization and gradually decreases when approaching the optimal solution, ensuring refined process parameter adjustment.

[0046] 5. By dynamically adjusting the process parameters in the gold ore refining process and setting reasonable constraints, it is ensured that the process parameters at every moment are within the reasonable range of the equipment carrying capacity and the refining reaction characteristics. This not only avoids equipment failure and out-of-control refining reaction, but also ensures smooth adjustment of process parameters in the optimization process, avoids drastic fluctuations, and helps to improve the safety and stability of the refining process. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a structural diagram of a gold ore refining intelligent control system according to the present invention;

[0048] Figure 2 This is a flow chart of the intelligent control method for gold ore refining described in the present invention. DETAILED DESCRIPTION

[0049] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0051] The specific scheme of a gold ore refining intelligent control system and method provided by the present invention is described in detail below with reference to the accompanying drawings.

[0052] Refer to the attached Figure 1, which shows a structure diagram of a gold ore refining intelligent control system provided by an embodiment of the present invention, the system includes the following parts:

[0053] Data acquisition module, data smoothing module, feature extraction module, gold recovery rate calculation module, process parameter optimization module, process parameter control module;

[0054] Data acquisition module: collects refining data in real time through the sensor system installed on the refining equipment, and outputs the refined data to the data smoothing processing module;

[0055] Data smoothing processing module: uses adaptive filtering and dynamic average processing algorithms to process the refined data from the data acquisition module to obtain the refined data after smoothing, and outputs the refined data after smoothing to the feature extraction module;

[0056] Feature extraction module: Based on the refined data after smoothing by the data smoothing processing module, the gold ore refining feature extraction algorithm is used to process the refined data to obtain the gold ore refining feature, and the gold ore refining feature is output to the gold recovery rate calculation module;

[0057] Gold recovery rate calculation module: The gold recovery rate is calculated by the gold recovery rate dynamic calculation algorithm based on the gold ore refining characteristics from the feature extraction module, and the gold recovery rate is output to the process parameter optimization module;

[0058] Process parameter optimization module: Based on the gold recovery rate of the gold recovery rate calculation module, the process parameters in the refining process are optimized using the gold ore refining process optimization algorithm to obtain updated values ​​of the process parameters, and the updated values ​​of the process parameters are output to the process parameter control module;

[0059] Process parameter control module: adjusts the actual process parameters according to the process parameter update values ​​of the process parameter optimization module.

[0060] Refer to the attached Figure 2 , which shows a flow chart of a gold ore refining intelligent control method provided by an embodiment of the present invention, the method comprising the following steps:

[0061] S1, collect refined data in real time, and process the refined data through adaptive filtering and dynamic averaging processing algorithm to obtain refined data after smoothing;

[0062] During the gold ore refining process, the sensor system installed on the refining equipment collects refining data in real time ,in, Indicated in The time point refining data, including but not limited to smelting temperature, ore particle size, moisture content and chemical composition, , Indicates the number of refined data;

[0063] The refined data collected by the sensor system in real time will contain noise and outliers. If not processed, it will affect the accuracy of subsequent analysis. Therefore, adaptive filtering and dynamic averaging processing algorithms are used to obtain smoothed refined data.

[0064] In order to reduce the impact of mutation points or short-term anomalies and keep the refined data in the refining process stable and continuous, the adaptive filtering and dynamic averaging processing algorithm not only considers the refined data at the current time point, but also considers the refined data at several time points in the past, i.e., historical refined data, through the dynamic averaging method, so as to calculate the average value of the refined data at each time point;

[0065] Adaptive filtering technology is further used to dynamically adjust the smoothing degree of the current refined data according to the difference between the refined data at each time point and the average value of the refined data, and the attenuation factor is used to control the influence of the historical refined data on the current refined data, ensuring that the influence of the historical refined data can be gradually weakened over time, thereby avoiding the situation where the historical refined data has too strong an impact on the current refined data processing, and the attenuation factor can also be dynamically adjusted according to different stages of the refining process;

[0066] By introducing a smoothness control parameter to adjust the smoothness of the refined data at each time point, it is possible to remove unnecessary noise and abnormal interference while maintaining the volatility and change trend of the refined data;

[0067] The calculation formula for the refined data after smoothing is:

[0068]

[0069] in, Indicated in After time point smoothing Refined data; Indicated in The time point Refined data; It means that the adjustment factor for smoothing is formed by performing power operation on the deviation value and weighting it; Indicated in The time point Refined data and running average The absolute difference between The control parameter representing the degree of smoothness is used to control the The intensity of the refined data smoothing process can be set according to the specific implementation scenario and is not limited here; Indicates The attenuation factor related to the refined data is used to control the effect of historical refined data on The time point The degree of influence of each refined data can be set according to the specific implementation scenario and is not limited here; Indicated in The time point The dynamic average of the refined data is calculated based on the historical refined data and used as the basis for smoothing. The time point The refined data is compared, and then the intensity of smoothing is adjusted to ensure the adaptability of the refined data. The calculation formula is:

[0070]

[0071] in, Indicates The sliding step size of the refined data is dynamically adjusted to reflect how much historical refined data is considered when calculating the dynamic average value at each time point; Indicates the time from To the time point within the time range of The refined data are summed; Indicated in The time point Refined data; Represents the normalization factor used to calculate the running average.

[0072] S2, based on the refined data after smoothing, using a gold ore refining feature extraction algorithm to obtain gold ore refining features;

[0073] Based on the refined data after smoothing, the gold ore refining feature extraction algorithm is used to obtain the gold ore refining features; the gold ore refining feature extraction algorithm uses a weighting coefficient to weight each refined data after smoothing, reflecting the degree of dependence of different refined features on different refined data, so as to control the importance of each refined data in feature extraction.

[0074] During the gold ore refining process, the refined data after smoothing may show periodic changes, and the periodicity can be simulated by the cosine function to capture the regularity of each smoothed refined data changing over time. The influence of time is taken into account so that the extracted refining features can reflect the periodic fluctuations in the gold ore refining process. Frequency parameters corresponding to different refining features are introduced to control the period of the cosine function and reflect the periodic change rate of different refining features.

[0075] In order to accurately extract the relationship between the refined data after smoothing, the phase of each refined feature is adjusted to ensure that the time series data are aligned, so as to accurately compare the dynamic changes of different refined features within the same time step, reveal the time dependency between different refined features, and eliminate the noise interference caused by phase differences;

[0076] The formula of gold ore refining feature extraction algorithm is:

[0077]

[0078] in, Indicated in The time point A refined feature; Represents the weighting coefficient, which is used to reflect the The refined feature pair The degree of dependence of each refined data is used to control the importance of each refined data in feature extraction. It can be set according to the specific implementation scenario and is not limited here. Indicated in After time point smoothing Refined data; Representation and time point The associated cosine function is used to capture the periodic changes in the refined data; Indicates The frequency parameter corresponding to the refined feature is used to control the period of the cosine function and reflect the The periodic change rate of a refined feature; Indicates The phase parameter corresponding to the refined feature determines the starting position or offset of the cosine function, which can help accurately locate the time point when the periodic change begins.

[0079] S3. Calculate the gold recovery rate by using a gold recovery rate dynamic calculation algorithm based on the gold ore refining characteristics;

[0080] The traditional gold recovery rate calculation method only considers the immediate impact of the characteristics on the gold recovery rate, but ignores the long-term impact of various refining characteristics on the gold recovery rate during the gold ore refining process. The gold recovery rate dynamic calculation algorithm uses a nonlinear weighted combination method for the gold ore refining characteristics, combining the immediate impact and cumulative effect of the refining characteristics to accurately calculate the gold recovery rate.

[0081] The influence of gold ore refining characteristics on gold recovery rate changes dynamically, and each refining characteristic needs to be weighted. The weight coefficient is set based on process experience, reflecting the immediate impact of each refining characteristic on gold recovery rate.

[0082] In addition to the immediate impact, the changes in some refining characteristics during the gold ore refining process have a hysteresis effect, which means that the impact of some refining characteristics on gold recovery is not immediate, but gradually appears over time. In order to accurately capture the long-term impact of refining characteristics, a cumulative effect term is introduced to reflect the cumulative effect of refining characteristics by integrating the changes in refining characteristics over a period of time.

[0083] The calculation formula for gold recovery is:

[0084]

[0085] in, Indicated in Gold recovery at the time point; Indicates the number of refined features; Indicates The weight coefficient of the refined feature reflects the The importance of each refining characteristic to gold recovery; Indicated in The time point A refined feature; Indicates The attenuation coefficient of each refining characteristic regulates the cumulative effect of the refining characteristic on gold recovery over a long period of time; is the cumulative effect term, indicating that The integration of each refining feature from the previous time point to the current time point reflects the cumulative effect of the refining feature and captures the impact of long-term dynamic changes on gold recovery rate; is the variable of integration, representing time; Indicated in The time point A refined feature.

[0086] S4. Based on the gold recovery rate, the process parameters are optimized using the gold ore refining process optimization algorithm to obtain updated values ​​of the process parameters, and the actual process parameters are adjusted according to the updated values ​​of the process parameters;

[0087] Based on the gold recovery rate, the gold ore refining process optimization algorithm is used to optimize the key process parameters in the refining process to achieve the goals of maximizing gold recovery rate and minimizing production costs.

[0088] The gold ore refining process optimization algorithm constructs an objective function, comprehensively considers the difference between the gold recovery rate and the target gold recovery rate and the difference between the production cost and the target production cost, and gradually optimizes the process parameters through an adaptive learning rate mechanism to achieve optimal control of the gold ore refining process.

[0089] The objective function calculates the difference between the gold recovery rate at the current time point and the target gold recovery rate, and the difference between the production cost and the target production cost, and combines the exponential decay term and the smoothness constraint of the process parameter changes to obtain the optimization target.

[0090] The difference between the current gold recovery rate and the target gold recovery rate and the difference between the production cost and the target production cost reflect the deviation between the current gold ore refining process and the target gold ore refining. An exponential decay term is further introduced to reflect the synergistic effect of the difference between the gold recovery rate and the target gold recovery rate and the difference between the production cost and the target production cost. As the difference decreases, the exponential decay term will gradually increase its influence on the optimization of the objective function, driving the optimization process to converge toward the optimal solution. By dynamically adjusting the weights of the difference between the gold recovery rate and the target gold recovery rate and the difference between the production cost and the target production cost in the objective function, the stability and convergence speed of the optimization process are further refined. The smoothing constraint ensures that drastic fluctuations are avoided when adjusting process parameters, thereby enhancing the stability of the gold ore refining process.

[0091] The objective function formula is:

[0092]

[0093] in, represents the objective function; Indicated in The difference between the gold recovery rate at a certain point in time and the target gold recovery rate is calculated using the formula: , Indicates the target gold recovery rate, which can be set according to the specific implementation scenario and is not limited here; Indicated in The difference between the production cost at a point in time and the target production cost is calculated as , Indicated in Production costs at a point in time, Indicates the target production cost, which can be set according to the specific implementation scenario and is not limited here; and The weight coefficients respectively represent the relative importance of adjusting the gold recovery rate difference and the production cost difference in the objective function; represents the exponential decay term, through which the synergistic effect between the difference in gold recovery rate and the difference in production cost is suppressed, thereby adjusting the sensitivity of the optimization process; Indicates the constant for controlling the attenuation strength, which can be set according to the specific implementation scenario and is not limited here; It represents the smoothness constraint term, which is used to prevent the drastic fluctuation of process parameters. By penalizing the drastic changes of process parameters, the adjustment of process parameters remains smooth. Indicated in The time point The value of a process parameter; Indicated in The time point The value of a process parameter; Indicates the number of process parameters; It represents the process parameter adjustment coefficient, which reflects the response intensity and sensitivity to the process parameter adjustment and is set according to expert experience.

[0094] The gold ore refining process optimization algorithm adopts an adaptive learning rate mechanism, which makes the learning rate extremely large in the early stage of optimization and can converge quickly, and gradually decreases when approaching the optimal solution, ensuring that the process parameters are carefully adjusted to obtain accurate optimization results;

[0095] The calculation formula for the updated value of the process parameter is:

[0096]

[0097] in, Indicated in The time point Updated values ​​of process parameters; Indicated in The time point The value of a process parameter; Indicated in The objective function at the time point is The gradient of the process parameters reflects the sensitivity of the objective function to the adjustment of the process parameters; Indicated in The learning rate at a time point can improve the optimization efficiency by dynamically adjusting the learning rate. The specific calculation formula is:

[0098]

[0099] in, Represents the initial learning rate, which is the learning rate value at the beginning of optimization. It can be set according to the specific implementation scenario and is not limited here; Represents the decay factor that controls the speed at which the learning rate decreases.

[0100] The process parameters in the gold ore refining process must follow a certain physical limit range, which is not only subject to the maximum load capacity of the equipment, but also subject to the chemical and physical characteristics of the refining reaction. In order to ensure that the updated values ​​of the process parameters at each time point are within a reasonable range, the following constraints are set:

[0101]

[0102] in, Indicates The minimum allowable value of each process parameter can be set according to the specific implementation scenario and is not limited here; Indicates The maximum allowable value of each process parameter can be set according to the specific implementation scenario and is not limited here;

[0103] The constraints ensure that during the optimization process, the process parameters will not exceed the limits that the equipment or process reaction can withstand, thereby avoiding equipment failure or loss of control of the refining reaction.

[0104] Finally, the actual process parameters are adjusted according to the updated values ​​of the process parameters.

[0105] In summary, an intelligent control system and method for gold ore refining has been completed.

[0106] The order of the embodiments of the invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0107] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0108] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. An intelligent control method for gold ore refining, characterized in that: The following steps are involved: S1: Collect refined data in real time, and process the refined data through adaptive filtering and dynamic averaging processing algorithm to obtain refined data after smoothing. The specific formula is: ； in, Indicated in After time point smoothing Refined data; Indicated in The time point Refined data; A control parameter indicating the degree of smoothing; Indicates The attenuation factor associated with each refined data; Indicates the calculation based on historical refined data The time point Dynamic average of refined data; S2: Using the gold ore refining feature extraction algorithm, each smoothed refined data is weighted by introducing a weighting coefficient, and the periodicity of the smoothed refined data is simulated by a cosine function; at the same time, frequency parameters corresponding to different refining features are introduced to control the period of the cosine function to obtain the gold ore refining features; S3: Based on the gold ore refining characteristics, the gold recovery rate is calculated by the gold recovery rate dynamic calculation algorithm; the specific formula is: ； in, Indicated in Gold recovery at the time point; Indicates the number of refined features; Indicates The weight coefficient of the refined features; Indicated in The time point A refined feature; Indicates The attenuation coefficient of the refined features; is the cumulative effect term; is the variable of integration; Indicated in The time point A refined feature; S4: Based on the gold recovery rate, the process parameters are optimized using the gold ore refining process optimization algorithm to obtain updated values ​​of the process parameters, and the actual process parameters are adjusted according to the updated values ​​of the process parameters.

2. The gold ore refining intelligent control method according to claim 1, characterized in that: The S4 specifically includes: The gold ore refining process optimization algorithm constructs an objective function and uses an adaptive learning rate mechanism to optimize process parameters to achieve optimal control of the gold ore refining process.

3. The gold ore refining intelligent control method according to claim 2 is characterized in that: The S4 specifically includes: The objective function is constructed by calculating the difference between the gold recovery rate and the target gold recovery rate and the difference between the production cost and the target production cost, and combining the exponential decay term and the smooth constraint term; the objective function formula is: ； in, represents the objective function; Indicated in The difference between the gold recovery rate at a certain point in time and the target gold recovery rate is calculated using the formula: , represents the target gold recovery rate; Indicated in The difference between the production cost at a point in time and the target production cost is calculated as , Indicated in Production costs at a point in time, represents the target production cost; and The weight coefficients respectively represent the relative importance of adjusting the gold recovery rate difference and the production cost difference in the objective function; represents the exponential decay term; represents the constant that controls the attenuation strength; represents the smoothness constraint; Indicated in The time point The value of a process parameter; Indicated in The time point The value of a process parameter; Indicates the number of process parameters; Represents the process parameter adjustment coefficient.

4. The intelligent control method for gold ore refining according to claim 3 is characterized in that: The S4 specifically includes: The gold ore refining process optimization algorithm uses an adaptive learning rate mechanism to update the process parameters and obtain the updated values ​​of the process parameters; the calculation formula is: ； in, Indicated in The time point Updated values ​​of process parameters; Indicated in The objective function at the time point is The gradient of the process parameters; Indicated in The learning rate at this time step.

5. The gold ore refining intelligent control method according to claim 4, characterized in that: The S4 specifically includes: Constraints are set so that the updated values ​​of the process parameters follow physical limits that are governed not only by the maximum capacity of the equipment but also by the chemical and physical characteristics of the refining reactions.

6. An intelligent control system for gold ore refining, applied to the intelligent control method for gold ore refining as claimed in claim 1, characterized in that: Includes the following parts: Data acquisition module, data smoothing module, feature extraction module, gold recovery rate calculation module, process parameter optimization module, process parameter control module; Data acquisition module: collects refined data in real time and outputs the refined data to the data smoothing processing module; Data smoothing processing module: Adopts adaptive filtering and dynamic average processing algorithm to process refined data, obtains refined data after smoothing, and outputs the refined data after smoothing to the feature extraction module; Feature extraction module: Based on the refined data after smoothing, the gold ore refining feature extraction algorithm is used to obtain the gold ore refining feature, and the gold ore refining feature is output to the gold recovery rate calculation module; Gold recovery rate calculation module: calculates the gold ore refining characteristics through the gold recovery rate dynamic calculation algorithm to obtain the gold recovery rate, and outputs the gold recovery rate to the process parameter optimization module; Process parameter optimization module: Based on the gold recovery rate of the gold recovery rate calculation module, the process parameters in the refining process are optimized using the gold ore refining process optimization algorithm to obtain updated values ​​of the process parameters, and the updated values ​​of the process parameters are output to the process parameter control module; Process parameter control module: adjusts the actual process parameters according to the process parameter update values ​​of the process parameter optimization module.

Images