An intelligent control method for slurry discharge pressure for underground cemented filling

Abstract

The present invention relates to the field of intelligent control technology. Further, it relates to an intelligent control method for slurry discharge pressure for underground cemented filling, which includes the following steps: Step 1: According to the current working condition parameters, establish an expected normal slurry discharge pressure model that can reflect the current working conditions, and obtain the expected normal slurry discharge pressure; Step 2: Compare the real-time measured slurry discharge pressure with the expected normal slurry discharge pressure to determine whether there is an abnormal increase; Step 3: Set an abnormal time window, consider the time cumulative effect, evaluate the severity of the identified pressure abnormality, and obtain an abnormality severity index; Step 4: According to the abnormality severity index, automatically adjust the slurry discharge flow rate to the target slurry discharge flow rate by adjusting the speed of the control pump. The present invention significantly improves the safety guarantee level during the pipeline transportation process of underground cemented filling.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent control, and particularly relates to an intelligent control method for slurry discharge pressure for underground cemented filling. Background Art

[0002] During the exploitation of mineral resources, a large number of mined - out areas will be generated. The existence of these mined - out areas may not only trigger geological disasters such as surface subsidence and rock stratum movement, threatening the safety of the mine and its surrounding environment, but also limit the further efficient recovery of resources. As an important means for dealing with mined - out areas, the cemented filling technology mixes materials such as waste rock, tailings, cementitious materials (such as cement) and water in a certain proportion to form a slurry, which is transported through a pipeline system and filled into the mined - out area. After solidification, it forms a filling body with a certain strength, thus effectively supporting the surrounding rock, controlling ground pressure, managing the mined - out area, and enabling safe secondary recovery or treatment of solid waste, achieving the goals of green mines and sustainable development. Therefore, the cemented filling system is an indispensable key part in modern mine production.

[0003] The cemented filling system generally includes links such as slurry preparation, mixing, pumping and pipeline transportation. Among them, pipeline transportation is the core link for transporting high - concentration slurry, which usually exhibits non - Newtonian fluid characteristics, from the ground preparation station or underground transfer station to the designated mined - out area over a long distance and at high pressure. The discharge pressure, that is, the pressure at the pump outlet or a certain monitoring point in the pipeline, is a key operating parameter in the whole transportation process. It not only directly reflects the working state and energy consumption of the pump, but also is an important indicator of the slurry flow state in the pipeline and the system safety. Maintaining the discharge pressure within a normal and reasonable range is crucial for ensuring the continuity, safety and economy of the filling operation.

[0004] However, in actual underground cemented filling operations, the stable control of the discharge pressure faces many challenges, and the abnormal increase in pressure is one of the common risks. This abnormal increase may be caused by various factors, such as: fluctuations in the slurry ratio or changes in the water - cement ratio resulting in sudden changes in its rheological properties (yield stress, plastic viscosity); gradual deposition, scaling or solidification of the slurry on the inner wall of the pipeline, resulting in a reduced flow cross - section and an increase in pipeline roughness; local blockage of the pipeline due to the mixing of large particles or foreign objects; or even malfunctions of the pumping equipment itself. If the abnormal increase in the discharge pressure fails to be identified and processed in a timely and effective manner, a series of serious consequences may occur, including: overloading of the pumping system, a sharp increase in energy consumption; excessive pipeline stress, posing a risk of pipe burst and endangering the safety of underground personnel and equipment; emergency shutdowns, pipeline flushing or even disassembly and cleaning forced to deal with high pressure or pipe blockage, which not only seriously interrupts the filling operation, affects the production plan, but also causes slurry waste, water resource consumption, and may lead to permanent damage to the pipeline due to the solidification of the slurry during stagnation, resulting in huge economic losses and maintenance costs. Summary of the Invention

[0005] The main object of the present invention is to provide an intelligent control method for the discharge pressure for underground cemented filling, which significantly improves the safety guarantee level in the pipeline transportation process of underground cemented filling, enhances the automation degree and operation efficiency of the system, and reduces the downtime and economic losses caused by pressure problems.

[0006] To solve the above problems, the technical solution of the present invention is realized as follows:

[0007] An intelligent control method for the discharge pressure for underground cemented filling, comprising the following steps:

[0008] Step 1: According to the current working condition parameters, establish an expected normal discharge pressure model that can reflect the current working condition to obtain the expected normal discharge pressure;

[0009] Step 2: Compare the real-time measured discharge pressure with the expected normal discharge pressure to determine whether there is an abnormal increase;

[0010] Step 3: Set an abnormal time window, consider the time cumulative effect, evaluate the severity of the identified pressure abnormality to obtain an abnormal severity index;

[0011] Step 4: According to the abnormal severity index, automatically adjust the discharge flow rate to the target discharge flow rate by adjusting the rotational speed of the control pump.

[0012] Furthermore, the working condition parameters include: The real-time density of the slurry at the moment of slurry temperature under the unit of kg / m³; the vertical height difference between the pipeline outlet and the pump outlet , unit of m; the total length of the filling pipeline , unit of m; the inner diameter of the filling pipeline , unit of m; the absolute roughness of the pipeline inner wall , unit of m; the plastic viscosity at the reference temperature , unit of Pa·s; the flow activation energy of the viscosity , unit of J / mol; the yield stress at the reference temperature , unit of Pa;

[0013] ​​​​​​​​Further, the expected normal slurry discharge pressure model is as follows:

[0014] ;

[0015] wherein, is the expected normal slurry discharge pressure at time is the effective friction coefficient function; is the Reynolds number at time is the real-time plastic viscosity of the slurry at time , with the unit of Pa·s; is the flow state correction coefficient, with a value of ;

[0016] Further, the real-time plastic viscosity is as follows:

[0017] .

[0018] Further, the effective friction coefficient function is as follows:

[0019] .

[0020] Further, the Reynolds number is as follows:

[0021] ;

[0022] wherein, is the average flow velocity of the slurry at time , with the unit of m / s; .

[0023] Further, the judgment of whether there is an abnormal increase is as follows:

[0024] ;

[0025] wherein, is the abnormal state flag at time , when , it indicates normal; when , it indicates abnormal; is the real-time measured slurry discharge pressure at time ; represents the relative deviation threshold coefficient, with a value range of 0.1 to 0.3; the larger the value of , the greater the allowable relative error.

[0026] Further, the abnormal severity index is as follows:

[0027] ;

[0028] Among them, is the time integration variable; is the abnormal time window.

[0029] Furthermore, the target slurry discharge flow is:

[0030] ;

[0031] Among them, is the maximum value of the set abnormal severity index; is the maximum flow reduction factor, .

[0032] An intelligent control method for slurry discharge pressure for underground cemented filling of the present invention has the following beneficial effects:

[0033] First of all, by establishing an accurate physical model that can dynamically predict the normal slurry discharge pressure according to real-time working condition parameters, this method overcomes the limitations of relying on fixed thresholds or simple trend judgments in the prior art. This model-based dynamic benchmark can more accurately distinguish normal pressure fluctuations from abnormal increases indicating potential risks, greatly improving the accuracy and reliability of abnormal identification, effectively reducing the occurrence of false alarms and missed alarms, and enhancing the signal-to-noise ratio of the monitoring system.

[0034] Secondly, after identifying the pressure abnormality, this method does not adopt a simple binary processing, but introduces a quantitative evaluation mechanism for abnormal severity. By comprehensively considering the amplitude of the pressure exceeding the normal range and the continuous cumulative effect within a specific time window, the system can make a more detailed and reasonable judgment on the actual harm degree of the abnormality. This severity assessment enables subsequent countermeasures to be more accurate and appropriate, avoiding over-reaction to minor and short-term abnormalities and ensuring timely attention to continuous and serious abnormalities.

[0035] This method realizes automatic and closed-loop slow-release control based on abnormal severity. When detecting a pressure abnormality and evaluating its severity, the system no longer simply gives a passive alarm or a rough shutdown, but can automatically calculate an adjusted target slurry discharge flow and smoothly achieve flow regulation by controlling the pump speed. The adjustment amplitude is proportional to the evaluated abnormal severity, that is, "treating minor problems lightly and major problems seriously". This intelligent and progressive flow reduction measure aims to actively reduce the pressure in the pipeline, relieve abnormal conditions, effectively prevent the occurrence or deterioration of pipe blockage accidents, and avoid secondary risks such as water hammer effect and solidification of slurry in the pipeline caused by traditional emergency shutdowns. Description of the Drawings

[0036] Figure 1 It is a schematic structural diagram of the cultivated land image height analysis and allocation decision-making system provided by the embodiments of the present invention. Specific embodiments

[0037] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0038] Reference Figure 1 : An intelligent control method for slurry discharge pressure for underground cemented filling, comprising the following steps:

[0039] Step 1: According to the current working condition parameters, establish an expected normal slurry discharge pressure model that can reflect the current working conditions, and obtain the expected normal slurry discharge pressure;

[0040] Step 2: Compare the real-time measured slurry discharge pressure with the expected normal slurry discharge pressure to determine whether there is an abnormal increase;

[0041] Step 3: Set an abnormal time window, consider the time accumulation effect, evaluate the severity of the identified pressure abnormality, and obtain an abnormal severity index;

[0042] Step 4: According to the abnormal severity index, automatically adjust the slurry discharge flow rate to the target slurry discharge flow rate by adjusting the rotation speed of the control pump.

[0043] In modern mining engineering, underground cementing filling is a key ground pressure management and goaf treatment technology, and its importance is becoming increasingly prominent. It is not only related to the production safety of mines and affects the recovery rate of resources, but also plays a vital role in the long-term stability of the mining environment. One of the core links of the cementing filling process is to transport the high-concentration, non-Newtonian fluid slurry mixed in a specific ratio from the ground slurry station or underground transfer station to the goaf to be filled through a pipeline system. In this long-distance, high-pressure transportation process, the slurry discharge pressure is a core operating parameter, which directly reflects the state of the transportation system and the flow characteristics of the slurry. However, due to the fluctuation of slurry properties (such as changes in density, viscosity, and yield stress), wear or scaling of the inner wall of the pipeline, and potential blockage risks, the slurry discharge pressure often increases abnormally. If this abnormal increase is not discovered and effectively handled in time, it will lead to reduced transportation efficiency and increased energy consumption at the least, and may cause pipeline rupture, equipment damage, and even interruption of filling operations at the worst, bringing serious safety hazards and economic losses. Traditional filling monitoring methods often rely on the operator's experience and judgment and simple pressure threshold alarms. This method has a delayed response and is difficult to accurately distinguish between normal pressure fluctuations and real blockage precursors, and it is even more impossible to achieve automated risk mitigation. Therefore, developing a method that can intelligently identify abnormal slurry pressure and automatically release it has great theoretical significance and engineering application value for ensuring the safe, efficient and stable operation of underground cementing filling operations.

[0044] The first step of this method, which is also the basis of the entire intelligent identification and slow release system, is to establish a mathematical model that can accurately reflect the expected normal slurry discharge pressure under the current working conditions. The core idea of this step is that only by accurately knowing what state the system "should" be in under the current conditions can we effectively judge whether the actual state is "abnormal". The slurry transported by downhole cementing filling is usually a typical Bingham fluid or a non-Newtonian fluid with similar rheological properties, and its flow behavior is far more complex than that of Newtonian fluids such as water. The magnitude of slurry discharge pressure is affected by a combination of factors, mainly including the hydrostatic pressure required to overcome gravity, the pressure required to overcome the internal yield stress of the slurry to start and maintain flow, the pressure loss along the way caused by the viscous flow of the slurry, and the pressure loss caused by the friction between the slurry and the inner wall of the pipeline. Therefore, a reliable expected normal slurry discharge pressure model must be able to comprehensively and accurately quantify the impact of these factors.

[0045] The model requires a series of real-time or quasi-real-time operating parameters, which jointly define the current conveying conditions. Specifically, these parameters cover the physical and rheological properties of the slurry itself. For example, the real-time density of the slurry at a specific temperature at the current moment, which is the basis for calculating the hydrostatic pressure; the rheological parameters of the slurry, mainly the plastic viscosity and yield stress, which directly determine the magnitude of the flow resistance, and these parameters are usually sensitive to temperature. Therefore, it is necessary to consider the reference value at the reference temperature, the flow activation energy of the viscosity, and the influence of the real-time temperature, and to correct the viscosity value in real time through a relationship similar to the Arrhenius equation. Similarly, the yield stress may also change with temperature and time (such as hydration reaction), and it is also necessary to obtain or estimate it in real time.

[0046] In addition, the model also requires the geometric parameters of the pipeline system, including the total vertical height difference between the pipeline outlet and the pump outlet, which determines the magnitude of the hydrostatic pressure; the total length and inner diameter of the filling pipeline, which are the key dimensional parameters for calculating the frictional pressure loss along the way; and the absolute roughness of the inner wall of the pipeline, which affects the frictional loss in the turbulent flow state.

[0047] Finally, the model also requires the current operating parameter, that is, the real-time measured volume flow rate of the slurry, because the flow rate directly affects the flow velocity, and thus affects the viscous loss and frictional loss.

[0048] Combining these input operating parameters, it is expected that the normal slurry discharge pressure model calculates the expected normal slurry discharge pressure value at a certain moment. This expression usually consists of several parts: the first part is the static pressure term calculated based on the real-time density and vertical height difference; the second part is the pressure term required to overcome the yield stress, calculated based on the real-time yield stress, pipeline length and diameter; the third part considers the plastic viscosity of the slurry, real-time flow rate, pipeline length and diameter, and describes the viscous pressure loss in the laminar flow or transitional flow state, which may include a coefficient related to the flow regime; the fourth part uses an effective friction coefficient function, combined with the slurry density, average flow velocity (calculated from the flow rate and pipe diameter), pipeline length and diameter, to calculate the frictional pressure loss in the turbulent flow or full flow regime.

[0049] Among them, the effective friction coefficient itself is a complex function, which depends on a modified Reynolds number (specially used to describe the flow state of non-Newtonian fluids, and its calculation requires density, flow velocity, pipe diameter, real-time plastic viscosity and real-time yield stress) and the relative roughness of the pipeline (the ratio of the absolute roughness to the inner diameter), and usually adopts an approximate form such as the Colebrook-White equation (such as the Churchill equation) for calculation.

[0050] Through such a model established based on the principles of fluid dynamics, which can dynamically reflect the changes in operating conditions, the system can obtain an accurate "normal" pressure baseline that varies with time. The "intelligence" of this step lies in the model's ability to dynamically adjust the expected pressure value according to the real-time changing input parameters, rather than using a fixed threshold. Implementing this step requires precise sensors (for measuring flow rate, temperature, density, etc.), a reliable parameter database (for pipeline geometric parameters, slurry reference rheological parameters, roughness, etc.), and a processing unit capable of performing complex calculations in real time (such as a PLC or an industrial control computer).

[0051] After establishing a reliable baseline for the expected normal slurry discharge pressure, the second step of this method is to perform real-time comparison to determine whether there is an abnormal increase in pressure currently. This step is the core of anomaly identification. The principle is to compare the slurry discharge pressure value measured in real time by the pressure sensor with the expected normal slurry discharge pressure value calculated by the model at the same time point. In theory, if the system is operating perfectly normally and the model is perfect, the two should be very close. However, in actual operating conditions, there will always be certain measurement errors, model deviations, and normal operating fluctuations. Therefore, it cannot be simply considered that any difference represents an anomaly.

[0052] To distinguish normal fluctuations from potential danger signals, a reasonable tolerance range needs to be set. This method adopts a strategy of relative deviation threshold. Specifically, the system calculates the absolute difference between the real-time measured pressure and the expected normal pressure, and then determines whether this difference exceeds a specific percentage of the expected normal pressure. This percentage is defined by a parameter called the "relative deviation threshold coefficient", and its value range is usually between 0.1 and 0.3 (i.e., 10% to 30%). The advantage of choosing a relative threshold instead of an absolute threshold is that it can adapt to the judgment under different working pressure levels. For example, during low-pressure transportation, a small absolute pressure deviation may be significant, while during high-pressure transportation, the same absolute deviation may still be within the normal range. The relative threshold enables the judgment criterion to automatically adapt to the current pressure baseline. The setting of the threshold coefficient is a key balance point: if it is set too small, the system will be too sensitive and may frequently give false alarms, judging normal fluctuations as anomalies; if it is set too large, it may miss alarms and fail to detect real early anomaly signals in a timely manner. Its specific value needs to be debugged and optimized according to the actual operating conditions, slurry characteristics, pipeline conditions, and the tolerance of risks. When the calculated absolute pressure deviation exceeds the relative tolerance range defined by the threshold coefficient, the system determines that there is an abnormal increase in pressure at the current moment and generates an anomaly status flag (for example, setting an internal variable to 1 to indicate an anomaly; setting it to 0 to indicate normal). Implementing this step depends on high-precision real-time pressure sensors and a control system capable of quickly performing comparison operations and logical judgments.

[0053] Merely detecting momentary pressure anomalies may not be sufficient to fully assess the risk. Sometimes, brief pressure spikes may be caused by temporary and self-recovering factors, while continuous and gradually accumulating pressure increases often indicate more serious problems, such as the gradual scaling on the inner wall of the pipeline or the formation of a slowly growing blockage. Therefore, the third step of this method introduces the concept of "abnormality severity assessment", aiming to quantify the severity of the abnormality and particularly considering the time cumulative effect. The principle of this step is that one should not only look at the current abnormal state but also review the pressure performance over a past period of time to comprehensively evaluate the amplitude and persistence of the abnormality. To this end, the system sets an "abnormality time window", which is a predefined time length (for example, it can be from a few minutes to more than ten minutes). When the second step detects an abnormality at the current moment (i.e., the abnormality status flag is 1), the system will initiate the severity assessment calculation. It will examine all the pressure data from the current moment back to the start of this time window. At each time point (or sampling point) within the time window, the system calculates how much the measured pressure exceeds the "allowed upper limit of normal fluctuation" (i.e., the expected normal pressure multiplied by (1 + relative deviation threshold coefficient)). This exceeded amount, if negative (i.e., the measured pressure is higher than the expected value but still within the tolerance, or lower than the expected value), is counted as 0; if positive, it means the pressure truly exceeds the allowed normal range. Then, the system integrates or averages this "exceeded amount" (usually normalized with respect to the expected normal pressure to obtain a relative exceeded ratio) over the entire abnormality time window. The result finally calculated is called the "abnormality severity index" (SI). This index comprehensively reflects the average relative degree to which the pressure anomaly exceeds the normal range over a past period of time.

[0054] A higher severity index means that within the recent time window, the pressure is not only significantly high but also this high state persists. On the contrary, if there are only brief and small-amplitude pressure spikes, even if they trigger an instantaneous anomaly, the severity index averaged over the time window will not be very high. The length of the abnormality time window is an important adjustment parameter: a shorter time window can more quickly reflect the cumulative effect of continuous anomalies but has a weaker smoothing effect on short-term fluctuations; a longer time window can better smooth short-term fluctuations and highlight long-term trends but may have a slightly slower response speed. The implementation of this step requires storing the pressure data (including measured values and model prediction values) of the recent period in the control system and having the ability to perform integration or moving average calculations. The introduction of the severity index enables the system to no longer be a simple "normal / abnormal" binary judgment but to quantify and classify the abnormal state, providing a basis for subsequent intelligent mitigation.

[0055] After the previous steps have successfully identified a pressure anomaly and evaluated its severity, the fourth step of this method, which is also the key link to ultimately achieve closed-loop control and actively mitigate risks, is to automatically adjust the slurry discharge flow rate according to the anomaly severity index. The core principle is that in most cases, reducing the flow velocity of the slurry in the pipeline helps to reduce the flow resistance (including viscous loss and frictional loss), thereby reducing the slurry discharge pressure. At the same time, for some initially formed soft blockages, reducing the flow velocity and pressure can sometimes cause them to loosen on their own or slow down their development. More importantly, this adjustment should be intelligent and adaptive, that is, the adjustment amplitude should match the severity of the problem, avoiding over-reaction or under-reaction in a "one-size-fits-all" manner. When the system confirms the existence of an anomaly (the anomaly status flag is 1) and calculates the anomaly severity index it will start the calculation program for the target slurry discharge flow rate. The system will pre-set a "maximum anomaly severity index" , which represents the severity benchmark that the system designer believes requires the most drastic braking measures. At the same time, a "maximum flow reduction factor" will also be set. This is a value between 0 and 1 (but not including 0), which defines the maximum allowable flow reduction ratio in a single automatic adjustment (for example, means that the flow rate will be reduced to at most 50% of the original flow rate).

[0056] The target slurry discharge flow rate is calculated as follows: First, calculate the ratio of the current severity index to the maximum reference value , that is . This ratio reflects the relative severity of the current anomaly. Then, multiply this ratio by the maximum flow reduction factor to obtain a preliminary reduction ratio based on severity. To prevent this ratio from exceeding the set upper limit (for example, even if the severity index exceeds the maximum reference value, the reduction ratio cannot exceed ), the system will take the smaller value between this preliminary reduction ratio and 1. Finally, subtract this final determined reduction ratio from 1, and then multiply by the current real-time slurry discharge flow rate , to obtain the adjusted target slurry discharge flow rate . This calculation process ensures that: when the severity index is low, the flow reduction amplitude is also small; as the severity index increases, the flow reduction amplitude gradually increases; when the severity index reaches or exceeds the maximum reference value, the flow reduction amplitude reaches the set maximum value . The calculated target slurry discharge flow rate It will be sent as an instruction to the control system of the pulp discharge pump. For example, by adjusting the output frequency of the variable frequency drive (VFD), the pump speed can be changed to precisely adjust the actual pulp discharge flow rate to the target value. This automated, severity-based flow regulation mechanism achieves intelligent and progressive relief of pressure anomalies, aiming to control the pressure within a safe range, win processing time for operators, or directly defuse risks in some cases, avoiding emergency shutdown of the system due to sudden high pressure. The implementation of this step requires the pump system to have adjustable speed capabilities (such as being equipped with a VFD), and the control system to be able to receive the target flow instruction and close-loop control the operation of the pump.

[0057] Furthermore, the operating conditions parameters include: The real-time density of the slurry at the moment of slurry temperature in , with the unit of kg / m³; the vertical height difference between the pipe outlet and the pump outlet , with the unit of m; the total length of the filling pipe , with the unit of m; the inner diameter of the filling pipe , with the unit of m; the absolute roughness of the pipe inner wall , with the unit of m; the plastic viscosity at the reference temperature in , with the unit of Pa·s; the flow activation energy of the viscosity , with the unit of J / mol; the yield stress at the reference temperature in , with the unit of Pa; The real-time volume flow rate of the slurry at the moment , with the unit of m³ / s; The real-time yield stress of the slurry at the moment of temperature in .

[0058] The real-time density of the slurry, denoted by , is a key factor in calculating the hydrostatic pressure component in the discharge pressure and also affects the calculation of pressure loss related to fluid inertia. For example, in a typical cement-tailings cemented filling ratio, the slurry density may be around kg / m³. This value can be obtained through real-time measurement by an online densitometer or estimated based on the ratio and component densities. Although the influence of temperature on density is relatively small, it may also need to be considered in a high-precision model. An accurate density value is crucial for calculating the pressure caused by the vertical height difference. Then comes the rheological properties of the slurry, which are particularly important for non-Newtonian filling slurries. The real-time yield stress, denoted by , with the unit of Pascal ( ) represents the minimum shear stress required to initiate slurry flow. It is one of the key characteristics differentiating the fill slurry from ordinary fluids and makes a significant contribution to the total pressure loss, especially at low flow rates or during the startup phase. For example, a slurry of medium consistency may have a yield stress of around Pa. However, this value is significantly affected by the mix ratio, solid content, type and dosage of admixtures, hydration time, and temperature, and may vary between Pa (thinner) and Pa (thicker or approaching initial setting). Therefore, it is crucial to obtain or estimate the yield stress at the current temperature in real time. This typically requires combining the yield stress at the reference temperature obtained from laboratory rheological tests ( ) and making corrections considering temperature and possible time-varying effects.

[0059] Equally important as the yield stress is the plastic viscosity of the slurry, which reflects the ability of the slurry to resist deformation internally after it starts to flow by overcoming the yield stress. The real-time plastic viscosity plays a decisive role in the viscous term of the pressure loss. Similar to the yield stress, the plastic viscosity also strongly depends on the slurry composition and temperature. In this method, the calculation of the real-time plastic viscosity is usually based on a reference plastic viscosity value measured at a reference temperature (for example, measured at 20 degrees Celsius and may be Pa⋅s) and the flow activation energy of the viscosity (for example, may be J / mol). The flow activation energy characterizes the sensitivity of the viscosity to temperature and is a parameter determined through rheological experiments at different temperatures. Using these parameters, the actual plastic viscosity at the current temperature can be calculated through a relationship similar to the Arrhenius equation (an empirical formula describing the variation of reaction rate or physical properties with temperature), combined with the real-time measured slurry temperature . For example, if the slurry temperature rises from the reference 20 degrees Celsius to 30 degrees Celsius, its plastic viscosity usually decreases due to the increased molecular thermal motion; conversely, if the temperature drops to 10 degrees Celsius, the viscosity will increase. Accurately considering this temperature dependence is crucial for maintaining the accuracy of pressure prediction under different ambient temperatures or when the slurry temperature changes due to hydration heat.

[0060] Secondly, the geometric parameters of the pipeline system form the "hardware" basis in the pressure model. The vertical elevation difference of the pipeline outlet relative to the pump outlet, in meters (m), directly determines the magnitude of the hydrostatic pressure. If the slurry is pumped downward (for example, from the surface to a deep underground goaf), is negative (for example m), at this time gravity helps the flow and reduces the outlet pressure requirement of the pump; if it is transported upward or horizontally and then upward, is positive (such as m), the pump needs to overcome this part of the gravitational potential energy difference additionally. The total length of the filling pipeline , in meters (m), is the basis for calculating the frictional pressure loss along the way (including viscous loss and frictional loss). The longer the pipeline, the greater the accumulated pressure loss naturally. For a long-distance filling system, the pipeline length may reach m or even longer. The inner diameter of the filling pipeline , in meters (m), has a very significant impact on the pressure loss. Since the pressure loss is usually inversely proportional to the fourth or fifth power of the pipe diameter, even a small change in the pipe diameter will result in a huge difference in the pressure requirement. For example, when transporting m³ / h of flow rate, using a pipeline with an inner diameter of mm compared to using a mm pipeline, its flow velocity will decrease significantly and the pressure loss will also be greatly reduced. Therefore, accurately knowing the inner diameter of the pipeline is the key to the accuracy of the model. The absolute roughness of the inner wall of the pipeline , in meters (m), describes the unevenness of the pipe wall surface. It mainly affects the frictional pressure loss at higher flow velocities (turbulent state). The roughness of a new steel pipe may be around mm, but with wear or scaling during use, the roughness may gradually increase, for example, increasing to mm or higher. In the pressure model, the relative roughness ( ) is usually used in combination with the Reynolds number to calculate the friction coefficient, and then the frictional pressure loss is obtained.

[0061] Finally, among the current operating parameters, the most core one is the real-time volume flow rate of the slurry at a certain moment , in cubic meters per second m³ / s. The flow rate directly determines the average flow velocity of the slurry in the pipeline (the flow velocity is equal to the flow rate divided by the cross-sectional area of the pipeline), and the flow velocity is a key variable affecting the viscous pressure loss and the turbulent frictional pressure loss. The flow rate is usually monitored in real time by a flow meter installed on the pipeline. For example, for a medium-scale filling operation, its flow rate may be set to m³ / s, which is converted to the standard unit as m³ / s. This real-time flow rate value is not only an important input for calculating the expected normal pressure, but also the target parameter to be adjusted during intelligent slow release in the fourth step.

[0062] Furthermore, the expected normal slurry discharge pressure model is:

[0063] ;

[0064] wherein, is the expected normal slurry discharge pressure at time is the effective friction coefficient function; is the Reynolds number at time is the real-time plastic viscosity of the slurry at time, with the unit of Pa·s; is the flow state correction coefficient, with a value of ;

[0065] The hydrostatic pressure caused by gravity is the first consideration in the model. In the expression, this part is usually reflected by the real-time density of the slurry , the acceleration due to gravity , and the vertical height difference between the pipe outlet and the pump outlet. The product of the three. Its physical meaning is to quantify the effect of the earth's gravity on the entire slurry column. When the filling slurry needs to be pumped to a place higher than the pump station, is positive, and the pump must provide additional pressure to overcome the resistance generated by the weight of the slurry itself, which is equivalent to "lifting" the entire vertical section of the slurry. Conversely, if the slurry is pumped downward to the goaf in the deep part of the mine, is negative. At this time, the gravity actually helps the slurry to flow and will generate a "siphon" effect, thus reducing the total pressure required at the pump outlet. Therefore, accurate density measurement and accurate pipeline elevation data are the keys to calculating this part of the pressure contribution, which is directly related to the static part of the energy balance.

[0066] Next, the model must deal with the influence brought by the rheological properties unique to the cemented filling slurry, especially the yield stress it has as a non-Newtonian fluid. Different from Newtonian fluids such as water or air, the filling slurry usually requires an initial driving force (shear stress) to overcome its internal structural strength before it can start to flow. This critical stress is the yield stress. In pipeline flow, this means that even in a horizontal pipeline, a certain pressure is required to maintain the overall flow state of the slurry and prevent it from stagnating or forming "plug flow". There is usually a pressure loss term related to the yield stress in the model, and its form may be similar to . This term indicates that the pressure required to overcome the yield stress is proportional to the total length of the pipeline , because the slurry throughout the length needs to be driven; it is proportional to the real-time yield stress , the thicker the slurry, the higher the initial threshold for resisting flow and the greater the required pressure; and it is related to the inner diameter of the pipeline Is inversely proportional because in a thinner pipe, a greater pressure gradient is required to generate sufficient shear stress at the pipe wall to overcome the yield stress. This part of the pressure loss is an important characteristic that differentiates the transportation of filled slurry from that of ordinary fluids. Especially at low flow rates or when the system restarts, its proportion can be quite significant.

[0067] After the slurry starts to flow, its internal viscosity is another important source of flow resistance. The model includes a part that describes this viscous pressure loss, which usually has a form related to the real-time plastic viscosity of the slurry , real-time volume flow rate , pipe length and the fourth power of the inner diameter of the pipe and may be written as . Here, the plastic viscosity refers to the viscosity exhibited by the slurry when it flows after overcoming the yield stress. This term indicates that the viscous pressure loss is proportional to the plastic viscosity (the more viscous the slurry, the greater the internal friction and the greater the loss), proportional to the volume flow rate (the faster the flow rate, the higher the shear rate, and usually the greater the viscous loss), and proportional to the pipe length (the longer the flow distance, the greater the cumulative loss). It should be particularly noted that the viscous pressure loss is inversely proportional to the fourth power of the inner diameter of the pipe , which means that a small change in the pipe diameter will have a huge impact on this part of the pressure loss, and the viscous resistance of a thin pipe is much greater than that of a thick pipe.

[0068] Finally, when the slurry flow rate is high, the flow state may change from a smooth laminar flow to a turbulent flow that contains a large number of vortices and mixing. In the turbulent flow state, the main mechanism of energy loss changes to the intense friction between fluid particles and the pipe wall and the turbulent dissipation within the fluid. The term used in the model to describe this part of the pressure loss usually adopts the structure of the Darcy - Weisbach formula, in the form of . Here, is the average flow velocity of the slurry (calculated from the flow rate and the pipe diameter ), represents the dynamic pressure of the fluid, that is, the kinetic energy per unit volume of the fluid. This dynamic pressure term is multiplied by the ratio of the pipe length to the diameter ( ), and then multiplied by a key dimensionless coefficient - the effective friction coefficient , to obtain the pressure loss caused by turbulent friction. The effective friction coefficient itself is a complex quantity, and its magnitude depends not only on the flow state (through the modified Reynolds number is also determined by the relative roughness of the inner wall of the pipeline . This frictional term indicates that at high flow velocities, the pressure loss is very sensitive to the square of the flow velocity ( ), and is also comprehensively affected by the pipeline length, diameter, fluid density, and the roughness condition of the pipe wall and the flow regime.

[0069] By algebraically superimposing the above four parts - hydrostatic pressure, pressure required to overcome the yield stress, viscous pressure loss in the laminar flow state, and frictional pressure loss in the turbulent flow state - it is expected that the normal slurry discharge pressure model will be able to give a comprehensive and clearly physically meaningful theoretical prediction value based on the current operating condition parameters. The importance of this model to the entire intelligent identification and mitigation method lies in that it provides a dynamically changing "normal" reference system that conforms to the current physical conditions. When the actually measured pressure significantly deviates from this carefully calculated expected value, the system has sufficient reason to judge that an abnormal situation (such as a precursor to pipeline blockage) may have occurred, thereby triggering subsequent mitigation measures such as severity assessment and flow rate adjustment.

[0070] Furthermore, the real-time plastic viscosity is:

[0071] .

[0072] Furthermore, the effective friction coefficient function is:

[0073] .

[0074] First, discuss the calculation of the real-time plastic viscosity . Plastic viscosity is a key parameter that describes the magnitude of the flow resistance of Bingham fluids or materials with similar rheological behaviors (such as cemented fill slurries) after they start to flow by overcoming the yield stress. It is similar to the viscosity of Newtonian fluids, but only comes into play after the shear stress exceeds the yield stress. During the actual filling process, the temperature of the slurry will continuously change due to factors such as environmental heat transfer and exothermic hydration reactions, and temperature is an important factor affecting the viscosity of the slurry. Therefore, accurately predicting the pressure requires real-time assessment of the plastic viscosity at the current temperature. The core part of the provided expression for calculating the real-time plastic viscosity contains an exponential term, which formally draws on the idea of the Arrhenius equation widely used in physical chemistry to describe the dependence of rate constants or certain physical properties (such as viscosity here) on temperature. This exponential term usually contains the flow activation energy , the ideal gas constant , and the current absolute temperature (obtained by converting the Celsius temperature plus 273.15). The flow activation energy is an experimentally determined parameter that characterizes the sensitivity of viscosity to temperature changes and reflects the energy barrier required to overcome intermolecular forces for flow. Based on this principle, when the real-time temperature increases, the value of the exponential term decreases (assuming a positive activation energy), resulting in a calculated real-time plastic viscosity relative to the plastic viscosity at the reference temperature to decrease, which is in line with the general law that most fluids "expand with heat and contract with cold" and have better fluidity at higher temperatures. Conversely, viscosity increases as temperature decreases. Therefore, by introducing a reference viscosity (a reference value measured at a specific reference temperature) and an activation energy , this part of the formula can capture the main trend of the change of plastic viscosity with temperature.

[0075] However, it should be particularly noted that in the formula for calculating the real-time plastic viscosity given, in addition to the above temperature-dependent term based on the Arrhenius principle, there is also a factor multiplied by the real-time yield stress . From the standard rheological model, plastic viscosity and yield stress are usually regarded as two independent rheological parameters that jointly describe the constitutive relationship of materials and may both be affected by factors such as temperature, composition, and time. However, it is generally not considered that plastic viscosity is directly proportional to the current yield stress. The appearance of this direct product relationship in this formula may reflect a special empirical coupling relationship found for the specific cemented fill slurry studied or the result derived under specific conditions (such as a certain simplified model setting). If the calculation is strictly performed according to this formula, its physical meaning is that in addition to the influence of temperature on viscosity itself, any factor that causes the real-time yield stress to increase (such as the progress of hydration, increase in solid content, etc.) will also directly increase the real-time plastic viscosity in proportion. This strong coupling relationship needs to be verified for its applicability in combination with specific experimental data or theoretical background. Despite this potential particularity or doubt, the calculated real-time plastic viscosity will be used for calculating the viscous pressure loss term in the main pressure model and is also one of the key inputs for calculating the modified Reynolds number , and its accuracy directly affects the accuracy of the entire pressure prediction.

[0076] Next, analyze the calculation formula of the effective friction coefficient function . This friction coefficient is used to quantify the energy loss caused by the friction between the fluid and the pipe wall and the turbulent dissipation inside the fluid during pipe flow, especially when the flow enters the transitional or turbulent state. It is the core parameter for calculating the last term (turbulent / frictional pressure loss term) in the main pressure model . The provided calculation The formula is an explicit empirical approximation formula, which is structurally similar to the well-known Churchill equation in fluid mechanics. Such equations aim to provide a method for estimating the friction coefficient that can cover the entire range from laminar flow to fully turbulent flow and from smooth pipes to rough pipes without iteration, and it is an effective approximation to the classical Colebrook-White equation that usually requires iterative solution.

[0077] This formula explicitly expresses the effective friction coefficient as a function of two key dimensionless parameters: the modified Reynolds number and the relative roughness . The modified Reynolds number is defined to adapt to the characteristics of non-Newtonian fluids such as cemented fill slurries. It combines the effects of fluid density, flow velocity, pipe diameter, plastic viscosity, and yield stress, and is used to determine whether the current flow state is laminar flow, transitional flow, or turbulent flow. The relative roughness directly reflects the influence of the physical roughness of the inner wall of the pipe relative to the pipe size. The structure of the formula, especially the two parts and included inside the logarithmic term, as well as the external coefficients 0.25 and the exponent -2, are all empirical constants and relationship forms obtained by fitting a large amount of experimental data. The principle is that: at lower Reynolds numbers or in smooth pipes, friction is mainly affected by viscosity (reflected by ); as the Reynolds number increases and / or the pipe becomes rough, the inertial effect and the obstructive effect of the rough elements on the pipe wall become increasingly important. The term in the formula reflects the influence of the Reynolds number on the friction coefficient. Generally, as the Reynolds number increases, the friction coefficient decreases (in the smooth and transitional regions of turbulent flow); while the term reflects the influence of the relative roughness. The greater the roughness, the higher the friction coefficient, especially at high Reynolds numbers (entering the fully rough region, the friction coefficient may no longer change with the Reynolds number and is only determined by the relative roughness). This explicit formula combines these two effects through a clever mathematical form, providing an estimate of the friction coefficient that is accurate enough and convenient to calculate in engineering.

[0078] Furthermore, the Reynolds number is:

[0079] ;

[0080] where, is the average velocity of the slurry at time , with the unit of m / s; .

[0081] The Reynolds number is a crucial dimensionless parameter in fluid mechanics. It predicts the flow regime by comparing the inertial force with the viscous force (or more generally, the resistance force) in fluid flow, that is, to determine whether the flow tends to be a smooth and orderly laminar flow or a chaotic and irregular turbulent flow. This is crucial for accurately calculating the frictional pressure loss in pipe flow. However, the classical definition of the Reynolds number (usually , where is the Newtonian viscosity) was proposed for Newtonian fluids (such as water and air). The slurry used for cemented filling, due to its high concentration of solid particles and gelling materials, usually exhibits typical non-Newtonian fluid characteristics, especially the presence of a yield stress . This means that when the slurry is subjected to a shear force below its yield stress, it behaves like a solid and does not undergo macroscopic flow; only when the shear stress exceeds the yield stress does it start to flow like a liquid and exhibit a certain plastic viscosity . This unique rheological behavior makes it inaccurate to directly apply the standard Reynolds number formula to judge its flow regime. Therefore, a modified Reynolds number is needed.

[0082] The numerator of the formula is exactly the same as the numerator form of the standard Reynolds number. It represents the inertial force in fluid flow. Here, is the real-time density of the slurry, is the average flow velocity of the slurry in the pipe (calculated from the real-time volume flow rate and the inner diameter of the pipe , ), is the inner diameter of the pipe. Physically, the greater the density, the faster the flow velocity, or the larger the pipe size, the stronger the inertia of the fluid to maintain its original motion state (or resist state change), and the more likely it is to develop into an unstable turbulent state due to a small perturbation.

[0083] The denominator of the formula represents the combined force resisting flow. It makes a crucial modification to the denominator of the standard Reynolds number (which only includes the viscous force represented by the viscosity ) to adapt to the characteristics of non-Newtonian fluids with a yield stress. This denominator consists of two parts: The first part is the real-time plastic viscosity . It represents the internal friction or viscosity exhibited by the slurry after it starts to flow (i.e., after overcoming the yield stress). This part acts similarly to the viscosity in Newtonian fluids, hindering the relative motion of the fluid and tending to suppress the occurrence of turbulence and maintain the laminar state. The second part is . This term is specifically to account for the yield stress introduced due to the influence on the flow state. The yield stress represents the internal structural strength that the slurry must overcome to change from a static state to a flowing state. This additional resistance also has the effect of inhibiting flow and stabilizing laminar flow. This term has the dimension of viscosity (unit ), and can be understood as the "equivalent viscosity" or resistance term contributed by the yield stress. It should be noted that the magnitude of this term is not only related to the properties of the slurry itself (yield stress ), but also related to the pipe diameter and the real-time flow velocity . Specifically, the greater the yield stress , the greater the value of this term, indicating a stronger resistance; when the flow velocity is lower, the contribution of this term is relatively more significant, meaning that at low flow velocities, the influence of the yield stress is more prominent in stabilizing the flow and preventing the transition to turbulence; while when the flow velocity is very high, the value of this term in the denominator will become smaller, indicating that the inertial force far exceeds the yield stress effect at this time, and the influence of the yield stress on the flow state criterion weakens. The coefficient '6' usually comes from the theoretical analysis or semi-empirical derivation of a specific rheological model (such as the Bingham model) during flow in a pipe, and is used to quantify the contribution of the yield stress to the effective resistance.

[0084] Therefore, the physical meaning of the entire modified Reynolds number can be understood as: at moment, the ratio of the inertial force (molecular) in the slurry flow to the total resistance force (denominator, including the viscous force represented by the plastic viscosity and the equivalent resistance contributed by the yield stress). By calculating this ratio, the current flow state of the slurry can be judged. Compared with the simple calculation that only considers the plastic viscosity, this modified Reynolds number usually gives a smaller value than the standard Reynolds number (if only calculated using ) because the influence term of the yield stress is added to the denominator. This appropriately reflects the physical reality: fluids with yield stress are more difficult to reach the turbulent state than fluids without yield stress (even with the same plastic viscosity), their laminar flow regions are wider, and a higher flow velocity or stronger perturbation is required to transition to turbulence.

[0085] This real-time calculated modified Reynolds number is a key intermediate step. It does not directly appear in the final pressure expression , but is used as an input parameter for calculating the effective friction coefficient . As mentioned before, the effective friction coefficient is a key factor for calculating the turbulent friction pressure loss, and the value of strongly depends on the flow state (determined by Characteristics) and the relative roughness of the pipeline. Therefore, accurately calculating the modified Reynolds number that can reflect the non-Newtonian characteristics of the slurry , is the prerequisite for ensuring the accurate calculation of the effective friction coefficient , and further ensures the accuracy of the estimation of the turbulent pressure loss part in the entire expected normal slurry discharge pressure model . If the calculation is inaccurate (for example, ignoring the influence of the yield stress and using the standard Reynolds number), it may misjudge the flow state, resulting in a calculation deviation, and ultimately causing the normal pressure predicted by the model to deviate from the actual situation. This deviation will directly affect the judgment of pressure anomalies in subsequent steps, and may cause the system to misjudge normal pressure fluctuations as anomalies, or fail to promptly identify a real abnormal increase, thus weakening the effectiveness and reliability of the entire intelligent monitoring and mitigation system. Therefore, adopting such a definition of the modified Reynolds number that takes into account the influence of the yield stress is crucial for accurately simulating the complex flow behavior of the cemented fill slurry and ensuring the performance of the intelligent system.

[0086] Furthermore, the judgment of whether there is an abnormal increase is as follows:

[0087] ;

[0088] where is the abnormal state flag at time. When , it indicates normal; when , it indicates abnormal; is the real-time measured slurry discharge pressure measured at time; represents the relative deviation threshold coefficient, and its value range is from 0.1 to 0.3; The larger the value of

[0089] Furthermore, the abnormal severity index is:

[0090] ;

[0091] where is the time integral variable; is the abnormal time window.

[0092] First of all, the judgment of the abnormal state is carried out. Its fundamental goal is to judge at each moment whether the currently measured slurry discharge pressure significantly deviates from the normal benchmark predicted by the model . Directly comparing the magnitudes of the differences between the two is not entirely sufficient because the normal pressure of the system itself may fluctuate within a wide range. For example, during the startup phase or at low flow rates, the pressure reference is relatively low, and at this time, a small absolute pressure deviation may indicate a problem; while during stable operation at high flow rates and high pressures, the same absolute deviation may be completely within the normal fluctuation range. To solve this problem, this method adopts a more intelligent and adaptive judgment criterion, that is, to compare whether the relative deviation between the measured pressure and the expected pressure exceeds a preset threshold. When calculating specifically, first find the and the absolute value of the difference between them, and then compare it with a specific proportion of the expected normal pressure , that is . Here, is a key adjustment parameter, called the relative deviation threshold coefficient, and its value range (for example, from 0.1 to 0.3) determines the tolerance of the system to deviations. The smaller the value, the more sensitive the system is and it can detect smaller relative deviations; the larger the value, the wider the allowed normal fluctuation range, which can reduce false judgments caused by noise or short-term disturbances, but may delay the recognition of slowly developing abnormalities. This judgment method based on relative deviation enables the abnormal detection standard to dynamically adapt to the current operating pressure level and improves the robustness of the judgment. Just judging that the system is in an abnormal state( ) is not enough because the degree of abnormality may vary greatly. A short-lived and small-amplitude pressure spike and a long-lasting and large-amplitude continuous increase in pressure clearly carry different risks and require different countermeasures. Therefore, after confirming the occurrence of an abnormality, the third step is to conduct an assessment of the severity of the abnormality, with the aim of quantifying how "severe" this abnormality is. The abnormality severity index is designed for this purpose. Its calculation principle is based on the cumulative consideration of the degree to which the pressure exceeds the normal range over a past period of time.

[0093] When , the system will start the calculation of . This calculation focuses on a preset time window in the past. Within this time window (from to ), the system will examine the pressure situation at each moment . The core of the calculation is to determine how much the measured pressure exceeds the "allowed upper limit of normal fluctuation". This upper limit is defined as the expected normal pressure plus the allowed relative deviation, that is . Then, calculate the difference between the two This difference represents the absolute amount by which the measured pressure exceeds the normal tolerance range. To eliminate the influence of fluctuations in the pressure reference itself and obtain a relative measure, this absolute excess is divided by the expected normal pressure at that time to obtain a relative excess ratio. Additionally, since we are only concerned with the severity of "abnormal pressure increases", if the measured pressure does not exceed this upper limit (i.e., the difference is negative or zero), its contribution to severity is considered zero, which is achieved by taking the maximum value between this relative excess ratio and 0( ).

[0094] Next, the formula integrates this non - negative "relative overpressure" calculated at each moment over the entire time window . The role of time integration is the cumulative effect: if the pressure only exceeds the limit briefly and slightly, the integral value will not be large; but if the pressure continuously and significantly exceeds the allowable range, the integral value will accumulate significantly. Finally, this integral result is divided by the length of the time window to obtain the average relative degree to which the pressure exceeds the normal tolerance upper limit over the past time period, which is the anomaly severity index . The choice of the time window has an important impact on the dynamic characteristics. A shorter window can reflect recent severe conditions more quickly but may have larger fluctuations; a longer window can smooth out instantaneous fluctuations and reflect the long - term average trend but has a slower response speed. The value of

[0095] is a non - negative number, and its magnitude intuitively reflects the severity of recent pressure anomalies: the larger the value, the more serious the situation where the pressure has continuously and significantly deviated from the normal range over a certain period of time. Furthermore, the target discharge flow rate

[0096] is as follows:

[0097] where is the maximum value of the set anomaly severity index; is the maximum flow reduction factor, .

[0098] This adjustment action is conditional. As clearly indicated by the trigger condition of the formula "if " (where represents the abnormal state according to the aforementioned logical analysis), the calculation of flow adjustment is only initiated when the system determines that there is a pressure anomaly currently. When the system is operating normally( ) In the case of , the pulp discharge flow rate will maintain its original set value or follow the normal control logic, and this automatic slow-release mechanism will not intervene. This conditional trigger mechanism ensures that the system only intervenes when necessary, avoiding unnecessary disturbances to the normal operating state.

[0099] The core principle of flow rate adjustment is to implement flow rate reduction that is appropriate to the severity of the abnormality. The system does not adopt a simple "on / off" control logic (for example, reducing the flow rate to a certain fixed low value once an abnormality occurs), but rather strives to achieve a more refined and intelligent proportional regulation. The basic idea is that the more severe the abnormality, the greater the extent of flow rate reduction; if the abnormality is relatively minor, the reduction extent will also be relatively small. This proportional regulation logic is reflected in the complex expression for calculating the flow rate reduction factor.

[0100] The calculation process of this expression can be understood by decomposition. First, the severity index ratio plays a key measurement role. It compares the currently real-time evaluated abnormality severity index with a pre-set reference benchmark . represents the severity threshold that the designer or operator believes requires significant intervention measures. This ratio thus provides a standardized relative severity measurement that ranges from 0 (theoretically, if the abnormality is just triggered but is extremely small) and above (if reaches or exceeds ). When is small, this ratio is less than 1, indicating that the abnormality degree is not yet serious; when equals , the ratio is 1, indicating that the preset critical severity level has been reached.

[0101] Next, this relative severity measurement is multiplied by the maximum flow rate reduction factor . is a parameter that ranges between 0 and 1 (usually not 0), and it sets the expected flow rate reduction ratio when the system reaches the critical severity level ( ). For example, if is set to 0.5, it means that when the abnormality severity reaches , the system plans to reduce the flow rate by 50% of the current flow rate. This parameter essentially adjusts the overall strength or "gain" of the control response. A smaller value means that even when the severity is relatively high, the initial reaction of the system is relatively mild, and the flow rate reduction extent is limited; a larger value indicates that the system reaction is more radical, and when reaching The choice of this parameter needs to be carefully weighed and set based on the specific pipeline characteristics, slurry properties, pumping system capabilities and operational risk preferences.

[0102] Therefore, the product represents the initial traffic reduction ratio calculated based on the current severity level and the preset response intensity. yes half of is 0.5, then the initial reduction ratio is , which means a 25% reduction.

[0103] However, considering extreme cases, e.g. Probably far more than ,or If set too large, directly using the above product may result in a calculated reduction ratio exceeding 100%, which is physically impossible (flow cannot be negative). To prevent this and ensure the stability of the system, the formula introduces The function is used as a safety constraint. The function takes the smaller value between 1 and the initial reduction ratio calculated previously. This means that no matter The final reduction ratio used to calculate the adjusted traffic will never exceed 1 (i.e. 100%). This constraint ensures that in any case, the target traffic More importantly, it allows for extremely severe exceptions (i.e. When the flow rate is greater than or equal to 1, the flow rate is reduced to zero to completely stop the pump as a final safety measure.

[0104] Finally, by subtracting the final reduction ratio after constraint processing from 1, a flow maintenance coefficient (between 0 and 1) is obtained, and then multiplied by the current real-time slurry discharge flow , the target slurry discharge flow rate is calculated .this The value will then be transmitted to the control system of the slurry discharge pump (for example, variable frequency drive VFD), and the control system will adjust the pump speed accordingly so that the actual slurry discharge flow rate approaches the newly set target value. The fundamental purpose of this automatic flow adjustment process is to reduce pressure loss by reducing the flow velocity of the slurry in the pipeline. According to the principles of fluid dynamics, the viscosity term and turbulent friction term in the pressure loss are usually positively correlated with the flow velocity (or flow rate). Therefore, reducing the flow rate can usually effectively reduce the total slurry discharge pressure, thereby alleviating the abnormal increase in pressure, creating conditions for the loosening or dissolution of potential blockages, or at least delaying the further development of blockages, giving operators more time to judge and deal with it.

[0105] A specific underground cemented filling operation scenario is presented in the form of an example to show the operation process of this intelligent recognition and slow-release method. Suppose a filling system is transporting slurry to an underground goaf, and the total length of the pipeline is 1500 meters, and the inner diameter is 0.15 meters. Based on elevation measurement, the vertical height difference between the pipeline outlet and the pump station outlet is -100 meters, indicating that the slurry is generally transported downward, and gravity will play a certain auxiliary role. The absolute roughness of the pipeline inner wall is evaluated to be approximately 0.0001 meters.

[0106] The basic rheological parameters of the slurry are measured at a reference temperature of 20 degrees Celsius (i.e., 293.15 Kelvin): the reference plastic viscosity is 0.15 Pa·s, and the reference yield stress is 40 Pa. The flow activation energy of the slurry viscosity is obtained through experiments to be approximately 22000 joules per mole. The control parameters of the system are set as follows: the relative deviation threshold coefficient is set to 0.15, the abnormal time window is set to 180 seconds (3 minutes), the maximum reference value of the abnormal severity index is set to 0.5, and the maximum flow reduction factor is set to 0.4.

[0107] At a certain monitoring moment , the real-time working condition data collected by the system are: the slurry temperature rises to 28 degrees Celsius (i.e., 301.15 Kelvin), possibly due to hydration heat or environmental influence; the real-time density is 1850 kg / m³; the real-time yield stress slightly rises to 45 Pa; the current slurry discharge volume flow rate is stable at 72 m³ / h, that is, 0.02 m³ / s. At the same time, the real-time slurry discharge pressure measured by the pressure sensor installed at the pump outlet is 2.5 MPa (2,500,000 Pa).

[0108] Now, start to execute its steps.

[0109] The first step is to calculate the expected normal slurry discharge pressure . The system first calculates the average flow velocity , , substituting the values to get m / s. Then, it is necessary to calculate the real-time plastic viscosity at the current temperature 。The reference viscosity is corrected according to the standard Arrhenius relation (note that this is different from the formula with a yield stress term in the problem description, and the standard form is used here to obtain a physically more reasonable viscosity value), and it is calculated that is approximately Pa·s, which is lower than the reference value of 0.15 Pa·s, in line with the effect of increasing temperature. Then, the corrected Reynolds number is calculated, which comprehensively considers the inertial force and the total resistance including the plastic viscosity and yield stress effects, . Substituting all real-time parameters, it is calculated that is approximately 282.5. This value indicates that the current flow regime may be in the laminar or transitional flow region. Using this corrected Reynolds number and the relative roughness , the system calculates the effective friction coefficient through an approximate formula (such as the Churchill-like formula), and obtains to be approximately 0.1075. Finally, considering all factors, the expected normal slurry discharge pressure is calculated. This pressure consists of four parts: the contribution of hydrostatic pressure ( , approximately -1.815 MPa, and the negative sign indicates gravity assistance), the pressure required to overcome the yield stress ( , approximately 1.80 MPa), the pressure loss due to viscous flow (assuming , approximately 0.056 MPa), and the turbulent / friction pressure loss ( , approximately 1.27 MPa). Adding these four terms together, we get MPa, which is approximately 1.311 MPa (1,311,000 Pa). This is the normal pressure level predicted by the model under the current operating conditions.

[0110] In the second step, the real-time measured pressure is compared with the expected normal pressure to determine if there is an abnormal increase. The measured pressure is 2.5 MPa, and the expected normal pressure is 1.311 MPa. Calculate the absolute difference between the two, which is MPa. At the same time, calculate the judgment threshold, which is 15% of the expected pressure ( ), and the threshold is MPa. Since the absolute difference of 1.189 MPa is much greater than the threshold of 0.197 MPa, the system determines that there is a significant abnormal increase in pressure currently. Therefore, the abnormal status flag is set to 1 (logically inferred to indicate abnormality).

[0111] In the third step, an assessment of the severity of the identified pressure abnormality is carried out. Since , the system starts to calculate the abnormal severity index . This requires looking back at the past 180 seconds ( ) The pressure data within. For simplicity of the example, assume that within these 180 seconds, the pressure has been in a state similar to the current one, that is, the measured pressure has continuously remained around 2.5 MPa, while the normal pressure predicted by the model and the tolerance upper limit ( MPa) are also relatively stable. The system calculates the relative amount by which the measured pressure exceeds the tolerance upper limit at each moment, that is . Since this value is greater than 0, after taking the maximum value, it is still 0.757. Assume that this value has been constant within the past 180 seconds, then is the integral average of this value, that is . This severity index of 0.757 is significantly greater than 0, indicating that within the past 3 minutes, the pressure has continuously and substantially exceeded the normal range.

[0112] Step 4, automatically adjust the slurry discharge flow rate according to the anomaly severity index. Since and , the system activates the flow rate adjustment mechanism. The current flow rate is 0.02 cubic meters per second. The system will compare with the set maximum severity index to calculate the relative severity . Then multiply by the maximum flow rate reduction factor to obtain the preliminary reduction ratio . Apply the safety constraint , and the finally determined reduction ratio is 0.6056. Calculate the target slurry discharge flow rate cubic meters per second. Converted back to common units, it is approximately 28.4 cubic meters per hour. Therefore, the intelligent system will immediately send an instruction to the control system of the pump (such as the frequency converter) to set the target flow rate to approximately 28.4 cubic meters per hour and achieve this goal by reducing the rotational speed of the pump.

[0113] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent control method for slurry discharge pressure for underground cemented filling, characterized in that, It includes the following steps: Step 1: According to the current working condition parameters, establish an expected normal slurry discharge pressure model that can reflect the current working condition to obtain the expected normal slurry discharge pressure; The expected normal slurry discharge pressure model is: ； Among them, is the expected normal slurry discharge pressure at time the effective friction coefficient function; is the Reynolds number at time is the real-time plastic viscosity of the slurry at time , with the unit of Pa·s; is the flow state correction coefficient, with a value of ; is the average flow velocity of the slurry at time ; Step 2: Compare the real-time measured slurry discharge pressure with the expected normal slurry discharge pressure to judge whether there is an abnormal increase; Step 3: Set an abnormal time window, consider the time cumulative effect, evaluate the severity of the identified pressure abnormality to obtain the abnormal severity index; Step 4: According to the abnormal severity index, automatically adjust the slurry discharge flow rate to the target slurry discharge flow rate by adjusting the speed of the control pump; The working condition parameters include: The real-time density of the slurry at the slurry temperature at a certain moment , with the unit of kg / m³; the vertical height difference between the pipeline outlet and the pump outlet , with the unit of m; the total length of the filling pipeline , with the unit of m; the inner diameter of the filling pipeline , with the unit of m; the absolute roughness of the pipeline inner wall , with the unit of m; the reference temperature The plastic viscosity at , with the unit of Pa·s; the flow activation energy of the viscosity , with the unit of J / mol; the reference temperature The yield stress at , with the unit of Pa; The real-time volume flow rate of the slurry at a certain moment , with the unit of m³ / s; The real-time yield stress of the slurry at a certain moment at the temperature ; , ; Real-time plastic viscosity is as follows: ； Effective coefficient of friction function is as follows: 。 2. The intelligent control method for slurry discharge pressure for underground cemented filling according to claim 1, characterized in that Reynolds number is as follows: ； Among them, is the average flow velocity of the slurry at a certain moment , with the unit of m / s; .

3. The intelligent control method for slurry discharge pressure for underground cemented filling according to claim 2, characterized in that Judging whether there is an abnormal increase is: ； Among them, is the abnormal state flag at a certain moment. When , it indicates normal; when , it indicates abnormal; is the real-time measured slurry discharge pressure at a certain moment; represents the relative deviation threshold coefficient, and its value range is from 0.1 to 0.3; The larger the value of , the greater the allowable relative error.

4. The intelligent control method for slurry discharge pressure for underground cemented filling according to claim 3, wherein Abnormal severity index is as follows: ； Among them, is the time integration variable; is the abnormal time window.

5. The intelligent control method for slurry discharge pressure for underground cemented filling according to claim 4, characterized in that, Target discharge flow rate is as follows: ； Among them, is the maximum value of the set abnormal severity index; is the maximum flow reduction factor, .

Images