Sludge interference control system for substation water-collecting well and intelligent desilting and draining device

Abstract

The invention discloses a sludge interference control system for a water collecting well of a transformer substation and an intelligent desilting drainage device.The sludge interference control system comprises a monitoring module used for monitoring a data set in the water collecting well, and the data set comprises sludge thickness, turbidity, water level and equipment vibration in the water collecting well; the fusion module is used for fusing the data sets; the state evaluation module is used for evaluating the fused data; and the decision control module is used for obtaining a control decision about whether the sludge is stirred or not based on the evaluation result. Intelligentization and automation of dredging and drainage of the water collecting well are achieved, the working efficiency is improved, the maintenance cost is reduced, and safe operation of equipment is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of silt clearing and drainage, and in particular to a silt interference control system and an intelligent silt clearing and drainage device for a water collection well of a transformer substation. Background Art

[0002] In the power system, substations are key nodes for power transmission and distribution, and their operational reliability is crucial to the stability of the entire power grid. As a facility within a substation used to collect and discharge rainwater and surface runoff, maintaining the proper function of the water collection well is crucial to ensuring the safe operation of the substation. However, over the long term, rainwater flushes into the water collection well, drawing impurities such as dust and mud from the ground into the well. These impurities gradually settle at the bottom of the water collection well, forming silt. Furthermore, microbial activity and the decomposition of plant debris within the water collection well also produce organic matter, which mixes with the mud and sand, further exacerbating the siltation problem.

[0003] Traditionally, desilting and draining water collection wells relies primarily on submersible pumps, but these pumps have significant limitations when handling stagnant water containing large amounts of silt. Because their pump bodies and impellers are not optimized for silt removal, they are unable to effectively remove silt during the pumping process, leading to its continuous accumulation within the collection well. Over time, this accumulated silt gradually covers the pump, causing it to run dry. This not only prevents effective drainage but can also lead to damage from overheating.

[0004] The installation and maintenance of water pumps in sump wells is also complex and carries significant risks. Installers must enter the well, requiring specialized safety precautions. Furthermore, the confined space and humid environment make installation and maintenance difficult and time-consuming. If a pump fails, maintenance personnel must also perform a series of tedious operations, including draining the water, testing for harmful gases, clearing silt from the well, and disassembling and replacing the pump and piping. These operations are not only time-consuming and labor-intensive, but can also pose safety risks to maintenance personnel.

[0005] Therefore, in view of the problems existing in the desilting and drainage of traditional water collection wells, it is particularly urgent to develop an intelligent desilting and automatic drainage device for water collection wells. Summary of the Invention

[0006] In response to the above problems, the present invention provides a silt interference control system and an intelligent silt clearing and drainage device for a substation water collection well that can monitor silt in real time, automatically clear silt and drain water, and improve efficiency and safety.

[0007] The technical solution of the present invention is: a silt interference control system for a water collection well in a substation, comprising:

[0008] A monitoring module, configured to monitor a data set in the water collection well, wherein the data set includes silt thickness, turbidity, water level, and equipment vibration in the water collection well;

[0009] Fusion module, used to fuse data sets;

[0010] State evaluation module, used to evaluate the fused data;

[0011] The decision control module is used to make a control decision on whether to stir the sludge based on the evaluation result.

[0012] The monitoring module includes an ultrasonic ranging array, a turbidity sensor, a liquid level differential pressure sensor and a vibration sensor.

[0013] The ultrasonic ranging array is used to monitor the silt distribution in the water collection well;

[0014] The turbidity sensor is used to monitor the turbidity in the water collection well;

[0015] The liquid level differential pressure sensor is used to monitor the water level in the water collection well;

[0016] The vibration sensor is used to monitor the sludge mixer located in the water collection well.

[0017] The fusion module includes:

[0018] Data preprocessing unit, used to standardize and process outliers in the data set of the monitoring module;

[0019] A state prediction unit is used to use the system model to predict the state at the next moment and obtain a predicted value;

[0020] The measurement update unit integrates the measurement value and the predicted value of the monitoring module;

[0021] The state estimation unit is used to obtain the optimal state estimation value.

[0022] The system model includes a state prediction equation and a covariance prediction equation, wherein,

[0023] The state prediction equation is:

[0024] X(k|k-1)=AX(k-1|k-1)+BU(k)

[0025] The covariance prediction equation is:

[0026] P(k|k-1)=AP(k-1|k-1)A'+Q

[0027] Where X is the state vector, including water level, turbidity, silt thickness, and equipment vibration; A is the state transfer matrix; B is the control input matrix; U is the control input vector, including the start and stop of the sludge mixer; P is the covariance matrix; Q is the process noise covariance matrix; A' is the transposed matrix of matrix A; k represents the current time, and k-1 represents the previous time.

[0028] The state evaluation module includes a state evaluation function, specifically:

[0029] S(t)=w1×U(t)+w2×T(t)+w3×V(t)+w4×L(t)

[0030] Where S(t) is the state evaluation value, U(t) is the standardized value of ultrasonic ranging data, T(t) is the standardized value of turbidity data, V(t) is the standardized value of vibration data, L(t) is the standardized value of liquid level differential pressure data, and w1, w2, w3, and w4 are weight coefficients respectively.

[0031] The decision control module includes:

[0032] When S(t) ≥ threshold 1, the controller starts the sludge mixer;

[0033] When S(t)≤threshold 2, the controller does not start the sludge mixer.

[0034] Also included is a monitoring cycle module, the monitoring cycle module is used to set a monitoring time period, and the monitoring module is used to monitor according to the time period;

[0035] The monitoring cycle module includes a dynamic adjustment function, specifically:

[0036] T=T0×(1+α×ΔS)

[0037] Where T is the adjusted working cycle, T0 is the standard cycle, α is the adjustment coefficient, and ΔS is the rate of change of the state evaluation value.

[0038] The intelligent desilting and drainage device for the substation water collection well includes a water pump intelligent control system and a silt interference control system for the substation water collection well.

[0039] The water pump intelligent control system includes a water pump instruction generating unit and a water pump executing unit.

[0040] The water pump instruction generating unit is used to generate control instructions for the water pump execution unit;

[0041] The water pump execution unit includes pump A and pump B, and the pump A and pump B are respectively slidably connected to the well wall lifting frame of the water collection well, and the vibration sensor is also used to monitor pump A and pump B;

[0042] The water pump instruction generating unit and the water pump executing unit are respectively connected to a controller, and the controller is wirelessly connected to a mobile phone APP or a control center.

[0043] The water pump instruction generation unit includes an alternating running time balancing function, specifically:

[0044] L=(T1-T2) / (T1+T2)×100%

[0045] Where: T1 and T2 are the cumulative running time of the two pumps respectively, and L is the running time imbalance. If |L| is greater than the balance threshold, the system adjusts its operation so that the running time of pump A and pump B tends to be balanced.

[0046] It also includes an adaptive optimization control unit, which is connected to the controller.

[0047] The adaptive optimization control unit includes optimizing the water pump intelligent control system using a deep reinforcement learning model;

[0048] The state space S of the deep reinforcement learning model is: S = {water level, flow rate, energy consumption, vibration};

[0049] The action space A is: A = {start, stop, speed adjustment};

[0050] The reward function R is:

[0051] R = w5×(energy efficiency ratio) + w6×(life expectancy) + w7×(maintenance cost)

[0052] Where: w5, w6 and w7 are weight coefficients respectively.

[0053] It also includes a flood warning system.

[0054] The flood warning system includes a collection module and a warning module.

[0055] The acquisition module is used to obtain real-time rainfall and liquid level differential pressure sensor data;

[0056] The early warning module is used to issue an early warning based on the acquired data; the early warning module is connected to the mobile phone APP or the control center via wireless;

[0057] The early warning module includes establishing an early warning model and performing graded early warning.

[0058] Among them, the early warning model is:

[0059] P(flood)=σ(W·X+b)

[0060] Where P(flood) is the probability of flood occurrence, σ is the standard deviation, b is the reference value or threshold, X is the eigenvector, W is the weight matrix,

[0061] The graded warnings are:

[0062] When P(flood) ≥ the first threshold, a red warning is issued;

[0063] When the second threshold ≤ P(flood) < the first threshold, an orange warning is issued;

[0064] When the third threshold ≤ P(flood) < the second threshold, a yellow warning is issued.

[0065] During operation, the silt interference system monitors the data of the water collection well, integrates the data, and evaluates the status. Finally, it decides whether to start the silt mixer based on the evaluation results.

[0066] In this way, the silt situation in the water collection well can be understood in real time. When a lot of silt accumulates, the silt can be destroyed by turning on the silt mixer to facilitate subsequent dredging operations.

[0067] When the intelligent dredging and drainage device is working, it can execute unit actions through the water pump to carry out dredging when dredging is required; it can also start forced drainage action when the water level reaches the warning height.

[0068] The present invention realizes the intelligent and automatic desilting and drainage of the water collection well, improves work efficiency, reduces maintenance costs, and ensures the safe operation of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 This is the principle block diagram of the silt interference control system in the present invention.

[0070] Figure 2 This is the principle block diagram of the intelligent silt removal and drainage device in the present invention.

[0071] Figure 3 1 is a schematic diagram of the three-dimensional structure of the intelligent silt removal and drainage device provided by an embodiment of the present invention;

[0072] Figure 4 It is a structural diagram of a sludge mixer;

[0073] In the figure, 1 is a sludge mixer, 2 is a sludge mixer lifting frame, 3 is a water pump, 4 is a water pump lifting frame, 5 is a turbidity sensor, 6 is a liquid level differential pressure sensor, 7 is a vibration sensor, 8 is a lifting rope, 9 is a drain pipe, 10 is a driving device, 11 is a stirring shaft, and 12 is an impeller. DETAILED DESCRIPTION

[0074] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and are not intended to limit the scope of protection of the present invention.

[0075] Reference Figure 1 The present invention provides a silt interference control system for a water collection well in a substation, comprising:

[0076] A monitoring module, configured to monitor a data set in the water collection well, wherein the data set includes silt thickness, turbidity, water level, and equipment vibration in the water collection well;

[0077] Fusion module, used to fuse data sets;

[0078] State evaluation module, used to evaluate the fused data;

[0079] The decision control module is used to make a control decision on whether to stir the sludge based on the evaluation result.

[0080] The monitoring module includes an ultrasonic ranging array, a turbidity sensor 5, a liquid level differential pressure sensor 6 and a vibration sensor 7.

[0081] The ultrasonic ranging array is arranged at the bottom of the water collection well;

[0082] The turbidity sensor has at least two turbidity sensors spaced from top to bottom on the well wall of the water collection well;

[0083] The liquid level differential pressure sensor has two sensors, which are respectively arranged on the two side walls of the water collection well;

[0084] The vibration sensor is used to monitor the sludge mixer 1 located at the bottom of the water collection well.

[0085] The specific arrangement needs to be adjusted according to the actual situation on site. The specific arrangement of the present invention is as follows:

[0086] Ultrasonic ranging array: A 4×4 array with a spacing of 1 meter is arranged at the bottom of the well to cover the entire bottom area for comprehensive monitoring of silt distribution;

[0087] Turbidity sensor: one is placed at 1 / 3, 1 / 2 and 2 / 3 of the well depth to form a three-dimensional monitoring network;

[0088] Liquid level differential pressure sensor: two groups are arranged symmetrically on both sides of the well wall, 0.5 meters and 1.5 meters from the bottom of the well;

[0089] Vibration sensor: One is placed on the sludge mixer and one on each of the two water pumps to monitor whether the equipment is swaying or shaking during operation.

[0090] The multi-sensor fusion design in the present invention has the advantages of data complementarity, redundant design, and three-dimensional monitoring, which can improve system reliability and monitoring accuracy.

[0091] Ultrasonic ranging array (4×4 matrix): accurately measures the thickness and distribution of silt layers, forms a complete bottom hole scan map, and evaluates dredging results;

[0092] Turbidity sensor (3 measuring points): monitors water quality changes in real time and provides water quality data for the system to judge and control the start and stop of the sludge mixer;

[0093] Liquid level differential pressure sensor (2 sets): monitors water level changes and provides water level data for the system to judge and control the start and stop of the water pump;

[0094] Vibration sensors (3 measuring points): 1 to monitor the sludge mixer, 2 to detect water pumps to monitor stability during operation and achieve predictive maintenance.

[0095] The fusion module includes:

[0096] Data preprocessing unit, used to standardize and process outliers in the data set of the monitoring module;

[0097] A state prediction unit is used to use the system model to predict the state at the next moment and obtain a predicted value;

[0098] The measurement update unit integrates the measurement value and the predicted value of the monitoring module;

[0099] The state estimation unit is used to obtain the optimal state estimation value.

[0100] The system model includes a state prediction equation and a covariance prediction equation, wherein,

[0101] The state prediction equation is:

[0102] X(k|k-1)=AX(k-1|k-1)+BU(k)

[0103] The covariance prediction equation is:

[0104] P(k|k-1)=AP(k-1|k-1)A'+Q

[0105] Where X is the state vector, which includes four state variables: water level, turbidity, silt thickness, and equipment vibration; A is the state transition matrix, which describes how the system state changes over time (i.e., the natural evolution of each state, such as water level drop and silt deposition); B is the control input matrix, which describes how control inputs affect the system state (i.e., the control effect of the sludge mixer, such as increasing turbidity and reducing silt); U is the control input vector, which includes control variables such as [water pump start and stop, sludge mixer start and stop]; P is the covariance matrix, which represents the uncertainty of the state estimate; Q is the process noise covariance matrix, which represents the uncertainty of the system model (i.e., the random interference of each state). A' is the transposed matrix of matrix A, where k represents the current time and k-1 represents the previous time. For example, X(k|k-1) represents the prediction of the state at time k based on the information at time k-1.

[0106] During operation, the present invention predicts specific values ​​such as water level and silt volume through state prediction; and evaluates the reliability of these predicted values ​​through covariance prediction to determine whether drainage operations are required later. Both are indispensable and together constitute complete prediction information.

[0107] By implementing a fusion module, this system provides the following results: real-time water level and accurate water level trends; silt distribution and bottom-well silt thickness distribution maps; water quality and turbidity trends; and equipment status and operational status assessments. It offers the following advantages: 1. Improved accuracy: Multi-source data complements each other, reducing the impact of single sensor errors; 2. Enhanced reliability: Redundant design ensures stable system operation; 3. Good real-time performance: Rapidly processes multi-source data to support real-time decision-making; 4. Strong anti-interference: Effectively filters out noise and improves data quality.

[0108] The state evaluation module includes a state evaluation function, specifically:

[0109] S(t)=w1×U(t)+w2×T(t)+w3×V(t)+w4×L(t)

[0110] Where S(t) is the state evaluation value, U(t) is the standardized value of ultrasonic ranging data, T(t) is the standardized value of turbidity data, V(t) is the standardized value of vibration data, L(t) is the standardized value of liquid level differential pressure data, and w1, w2, w3, and w4 are weight coefficients respectively.

[0111] Setting a state evaluation function has the following advantages:

[0112] 1. Multi-dimensional evaluation: Comprehensively consider four key indicators: water level, turbidity, vibration, and liquid level differential pressure; avoid the limitations of single indicator judgment; and improve the comprehensiveness and accuracy of the evaluation;

[0113] 2. Reasonable weight distribution: The liquid level differential pressure weight (e.g. 0.4) is used as an auxiliary indicator to provide water level change information to automatically control drainage and realize the basic functions of the entire system;

[0114] The turbidity weight (e.g. 0.3) reflects the water quality. Based on the water quality, the system determines whether to start the sludge mixer before the next drainage.

[0115] The ultrasonic ranging weight (e.g. 0.2) reflects the silt distribution and determines the dredging effect;

[0116] The vibration weight (such as 0.1) reflects the operating status of the equipment and affects the stability of the system.

[0117] 3. Standardization processing: All input data are standardized to eliminate the influence of different dimensions and facilitate data comparison and fusion.

[0118] 4. Strong real-time performance: The calculation is simple and fast, supporting real-time status evaluation and facilitating rapid system response.

[0119] 5. Good scalability: The weight coefficient can be adjusted according to actual needs, and new evaluation indicators can be easily added to adapt to different application scenarios.

[0120] The decision control module includes:

[0121] When S(t) ≥ threshold one (e.g., 0.7), the controller starts the sludge mixer;

[0122] When S(t)≤threshold 2 (eg 0.3), the controller does not start the sludge mixer.

[0123] The sludge mixer in the present invention is arranged at the bottom of the water collection well, such as Figure 4 As shown in the figure, the sludge mixer mainly includes the following:

[0124] Impeller 12: The core component of the sludge mixer, responsible for producing the stirring effect. The design of the impeller directly affects the stirring effect and energy efficiency.

[0125] Driving device 10: a motor, used to drive the impeller to rotate.

[0126] Stirring shaft 11: connects the driving device and the impeller to transmit power to the impeller.

[0127] Bracket and base: used to fix the agitator to ensure stable operation of the equipment.

[0128] The sludge mixer of the present invention adopts a turbine-type impeller design. The device has a high rotational speed, can generate a strong vertical flow and a special waveform, and can disperse the sludge in the water collection well in all directions, thereby performing a jamming action. The sludge mixer is slidably connected to the well wall lifting frame of the water collection well.

[0129] Also included is a monitoring cycle module, the monitoring cycle module is used to set a monitoring time period, and the monitoring module is used to monitor according to the time period;

[0130] The monitoring cycle module includes a dynamic adjustment function, specifically:

[0131] T=T0×(1+α×ΔS)

[0132] Where T is the adjusted duty cycle, T0 is the standard cycle (20 minutes), α is the adjustment coefficient (range 0.1-0.5, default 0.3), and ΔS is the rate of change of the state evaluation value (the difference between the current state value and the previous cycle).

[0133] The time period design in the present invention is described as follows:

[0134] 1. The standard cycle (20 minutes / time) is set based on experience, which can balance the desilting effect and energy consumption and is suitable for general working conditions.

[0135] 2. Dynamic adjustment settings can adapt to different working conditions and adjust according to the silt accumulation rate; can optimize energy consumption and avoid unnecessary frequent silt removal; improve efficiency and remove silt in time when needed.

[0136] 3. Adjustment mechanism:

[0137] When ΔS>0: the state deteriorates and the cycle is shortened;

[0138] When ΔS<0: the state improves and the cycle is extended;

[0139] When ΔS=0: maintain the standard cycle.

[0140] 4. Set the adjustment coefficient α to control the adjustment range, prevent excessive periodic changes, and ensure system stability.

[0141] like Figure 2 As shown, the intelligent desilting and drainage device for the substation water collection well includes a water pump intelligent control system and a silt interference control system for the substation water collection well.

[0142] The water pump intelligent control system includes a water pump instruction generating unit and a water pump executing unit.

[0143] The water pump instruction generating unit is used to generate control instructions for the water pump execution unit;

[0144] The water pump execution unit includes pump A and pump B, and the pump A and pump B are respectively slidably connected to the well wall lifting frame of the water collection well, and the vibration sensor is also used to monitor pump A and pump B;

[0145] The water pump instruction generating unit and the water pump executing unit are respectively connected to a controller, and the controller is wirelessly connected to a mobile phone APP or a control center.

[0146] The two water pumps are respectively slidably connected to the lifting frame on the wall of the water collection well and installed at a position 20-30 cm away from the bottom of the water collection well. The specific installation height is determined by the depth and capacity of the water collection well.

[0147] The controller is located near the wellhead and can be mounted on the wall or on the floor. It is connected to the sensor and water pump through wired connections. It is wirelessly connected to the mobile phone APP and the control center to achieve remote monitoring and operation.

[0148] The water pump instruction generation unit includes an alternating running time balancing function, specifically:

[0149] L=(T1-T2) / (T1+T2)×100%

[0150] Where: T1 and T2 are the cumulative running time of the two pumps respectively, L is the running time imbalance, which indicates the degree of difference in the running time of the two pumps. When L = 0, it means that the running time of the two pumps is exactly the same. When L>0, it means that pump A runs more than pump B. When L<0, it means that pump B runs more than pump A. If |L| is greater than the balance threshold, the system adjusts its operation so that the running time of pumps A and B tends to be balanced.

[0151] Balance threshold: When |L|≤10%, it means the maximum allowed running time difference is 10%.

[0152] When |L|>10%, the system will automatically adjust the operation strategy to balance the operating time of the two pumps; the purpose of setting the threshold is to ensure that the wear degree of the two pumps is similar and extend the service life of the equipment.

[0153] Set up dual-pump coordinated control, with pumps A and B running alternately to facilitate maintenance and replacement; the specific rotation operation is: running pump A this time, and automatically switching to pump B next time to ensure operational reliability.

[0154] It also includes an adaptive optimization control unit, which is connected to the controller.

[0155] The adaptive optimization control unit includes optimizing the water pump intelligent control system using a deep reinforcement learning model;

[0156] The state space S of the deep reinforcement learning model is: S = {water level, flow rate, energy consumption, vibration};

[0157] The action space A is: A = {start, stop, speed adjustment};

[0158] The reward function R is:

[0159] R = w5×(energy efficiency ratio) + w6×(life expectancy) + w7×(maintenance cost)

[0160] Wherein: w5, w6 and w7 are weight coefficients, which can be set according to actual conditions, such as w5 = 0.4, w6 = 0.3, w7 = 0.3.

[0161] The specific optimization process is:

[0162] Optimization of the state space S:

[0163] Water level: collected and normalized in real time by sensors to determine the system operating status;

[0164] Flow rate: Normalized based on the pump characteristic curve to reflect the system workload;

[0165] Energy consumption: Calculate and normalize the energy consumption per unit time to evaluate system efficiency;

[0166] Vibration: Feature extraction and normalization of vibration signals for equipment health monitoring;

[0167] Optimization of action space A:

[0168] Startup: Make intelligent judgments based on water level thresholds and system status to ensure safe startup;

[0169] Stop: Make decisions based on safety thresholds and operating status to prevent abnormal shutdowns;

[0170] Speed ​​regulation: Dynamically adjust the speed according to load demand to achieve optimal operation.

[0171] in:

[0172] Energy efficiency ratio: reflects the operating efficiency of the system;

[0173] Life expectancy: Evaluate the equipment's service life;

[0174] Maintenance cost: Calculate system maintenance costs;

[0175] w5, w6, and w7 are weight coefficients used to dynamically adjust the optimization focus.

[0176] An adaptive optimization control unit is set up to improve energy utilization efficiency, extend equipment life, and reduce maintenance costs while ensuring drainage effects; enabling the system to automatically adjust according to actual operating conditions.

[0177] It also includes a flood warning system.

[0178] The flood warning system includes a collection module and a warning module.

[0179] The acquisition module is used to obtain real-time rainfall and liquid level differential pressure sensor data;

[0180] The early warning module is used to issue an early warning based on the acquired data; the early warning module is connected to the mobile phone APP or the control center via wireless;

[0181] The early warning module includes establishing an early warning model and performing graded early warning.

[0182] Among them, the early warning model is:

[0183] P(flood)=σ(W·X+b)

[0184] Where P(flood) is the probability of flood occurrence, ranging from 0 to 1. A larger value indicates a higher probability of flood occurrence. This is a probability value used to assess flood risk.

[0185] σ is the standard deviation, which is used to measure the degree of dispersion of data. The larger the value, the greater the data fluctuation.

[0186] In flood warning, it is used to assess the volatility of water level changes;

[0187] b is the baseline value or threshold, which is used to set the critical point of the early warning. When this value is exceeded, the early warning is triggered. It is usually set based on historical data and experience;

[0188] These three parameters together constitute a flood warning model. P(flood) is used to assess the overall risk, σ is used to assess the uncertainty of the data, and b is used to set specific warning standards. This model can predict the possibility of floods and improve the reliability of assessment and warning.

[0189] X is the feature vector (including rainfall, water level and other data), W is the weight matrix,

[0190] The graded warnings are:

[0191] When P(flood) ≥ the first threshold (e.g., 0.8), a red alert is issued;

[0192] When the second threshold (e.g., 0.6) ≤ P(flood) < the first threshold, an orange warning is issued;

[0193] When the third threshold (such as 0.4) ≤ P(flood) < the second threshold, a yellow warning is issued.

[0194] Specifically, the colors of the warning levels generally follow the following:

[0195] Red: The flood risk level is the highest, and the accumulated water will soon overflow the collection well.

[0196] Orange: The flood risk is secondary, both pumps are fully open, and the water level is still rising.

[0197] Yellow: Attention, both pumps are fully open, and the water level has not changed.

[0198] In application, the present invention can realize automatic silt removal, intelligent drainage, and remote monitoring. The specific installation process includes:

[0199] Lifting frame installation: First, accurately install the lifting frame at the predetermined position on the water collection well wall; ensure that the lifting frame is installed firmly and reliably to withstand the weight of subsequent equipment and the stress during operation.

[0200] Reference Figure 3 The overall structural diagram of the device shown is divided into a sludge mixer lifting frame 2 and a water pump lifting frame 4.

[0201] Installation of the sludge mixer: The sludge mixer 1 is made of corrosion-resistant stainless steel. Slide it onto the lifting frame and slowly lower it to the bottom of the sump using the lifting rope 8. Ensure that the sludge mixer maintains a good fit with the lifting frame during the descent and is stably fixed once it reaches the designated position.

[0202] Pump 3 Installation: Pumps A and B are similarly slidably connected to the lifting frame and installed using the lifting rope 8. During installation, ensure that the pump's descent path is smooth and unobstructed, and that the bottom of the lifting frame maintains an appropriate distance from the bottom of the water collection well after reaching the bottom of the water collection well. This distance should be adjusted according to the actual depth of the water collection well.

[0203] Connecting the water pump 3 to the drain pipe 9: All lifting cables connecting the equipment should be made of stainless steel to ensure strength and durability. The cables should be designed for easy manual operation to raise and lower the sludge mixer and water pump. The pump should also be equipped with a quick-connect drain pipe. The material can be selected based on actual application requirements. Flexible hoses offer improved flexibility and foldability, while rigid pipes utilize quick connectors for quick connection and disconnection.

[0204] Reference Figure 3 The overall structural diagram of the device shown is divided into a sludge mixer lifting cable and a water pump lifting cable.

[0205] Repair and replacement: The design of the lifting rope allows maintenance personnel to operate at the wellhead. By pulling the equipment up and down through the lifting rope, they can complete the repair and replacement of the equipment without entering the water collection well.

[0206] Operation and maintenance personnel can remotely view changes in water levels in water pumps, sludge mixers, and collection wells through the control center computer or app to determine the operating status of the equipment.

[0207] The sludge interference system of the present invention monitors the data of the water collection well, integrates the data, and evaluates the status. Finally, it decides whether to start the sludge mixer based on the evaluation results.

[0208] In this way, the silt situation in the water collection well can be understood in real time. When a lot of silt accumulates, the silt can be destroyed by turning on the silt mixer to facilitate subsequent dredging operations.

[0209] When the intelligent dredging and drainage device is working, it can execute unit actions through the water pump to carry out dredging when dredging is required; it can also start forced drainage action when the water level reaches the warning height.

[0210] The liquid level differential pressure sensor detects the water level. When no water flows into the water collection well, there is no water level change, and the water pump and sludge mixer are in the off state.

[0211] When the water level in the collection well continues to rise and the differential pressure sensor detects that the water level has reached the required pump activation level, the system determines based on real-time data whether the conditions for activating the sludge mixer are met. If the accumulated water quality reaches the point where sludge is generated, the system will first activate the sludge mixer to reduce sludge formation. Subsequently, the system continuously monitors turbidity changes. Once the turbidity reaches a peak, the system initiates a command to activate the forced drainage pump to remove the highly turbid wastewater.

[0212] On the contrary, if the water quality of the accumulated water is relatively clear and does not meet the conditions for silt generation, the system will skip starting the silt mixer and directly issue a command to start the strong drainage pump to efficiently discharge the accumulated water in the collection well, ensuring the operating efficiency and water quality control of the collection well.

[0213] The present invention can improve the siltation treatment capacity of the water collection well and reduce the generation of silt, thereby effectively improving the overall performance of the drainage system and reducing maintenance costs.

[0214] The present invention has the following significant advantages:

[0215] 1. High degree of intelligence

[0216] Fully automatic operation: The equipment starts and stops automatically without manual intervention; supports remote monitoring and operation management; and enables real-time monitoring of equipment operating status.

[0217] 2. Significantly improved operational efficiency

[0218] Silt treatment is more thorough, silt re-deposition is reduced, water pump operation efficiency is improved, and manpower input is greatly reduced.

[0219] 3. Significantly improved safety performance

[0220] It can realize operations without going down into the well, avoid direct contact between personnel and sewage, and provide intelligent flood warning.

[0221] 4. Significant environmental benefits

[0222] Reduce the risk of sewage overflow, improve water quality treatment effects, and reduce water resource waste.

[0223] 5. Significant social benefits

[0224] Improve drainage reliability, ensure safe operation of substations, and enhance power supply reliability; easy to promote and apply.

[0225] The present invention greatly improves the management level and operation efficiency of the water collection well of the substation, and provides a reliable guarantee for the safe and stable operation of the power system.

[0226] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above specific embodiments. Without departing from the core idea of ​​the present invention, those skilled in the art can still make various changes and adjustments to the present invention within the scope of their professional knowledge.

Claims

1. The silt interference control system for the substation water collection well is characterized by: include: A monitoring module, configured to monitor a data set in the water collection well, wherein the data set includes silt thickness, turbidity, water level, and equipment vibration in the water collection well; Fusion module, used to fuse data sets; The fusion module includes: Data preprocessing unit, used to standardize and process outliers in the data set of the monitoring module; A state prediction unit is used to use the system model to predict the state at the next moment and obtain a predicted value; The measurement update unit integrates the measurement value and the predicted value of the monitoring module; A state estimation unit, used to obtain an optimal state estimation value; State evaluation module, used to evaluate the fused data; The decision control module is used to make a control decision on whether to stir the sludge based on the evaluation result.

2. The silt interference control system for the substation water collection well according to claim 1 is characterized in that: The monitoring module includes an ultrasonic ranging array, a turbidity sensor, a liquid level differential pressure sensor and a vibration sensor. The ultrasonic ranging array is used to monitor the thickness of silt in the water collection well; The turbidity sensor is used to monitor the turbidity in the water collection well; The liquid level differential pressure sensor is used to monitor the water level in the water collection well; The vibration sensor is used to monitor the sludge mixer located in the water collection well.

3. The silt interference control system for the substation water collection well according to claim 1 is characterized in that: The system model includes a state prediction equation and a covariance prediction equation, wherein, The state prediction equation is: X(k|k-1)=AX(k-1|k-1)+BU(k) The covariance prediction equation is: P(k|k-1)=AP(k-1|k-1)A'+Q Where X is the state vector, including water level, turbidity, silt thickness, and equipment vibration; A is the state transfer matrix; B is the control input matrix; U is the control input vector, including the start and stop of the sludge mixer; P is the covariance matrix; Q is the process noise covariance matrix; A' is the transposed matrix of matrix A; k represents the current time, and k-1 represents the previous time.

4. The silt interference control system for the substation water collection well according to claim 1 is characterized in that: The state evaluation module includes a state evaluation function, specifically: S(t)=w1×U(t)+w2×T(t)+w3×V(t)+w4×L(t) Where S(t) is the state evaluation value, U(t) is the standardized value of ultrasonic ranging data, T(t) is the standardized value of turbidity data, V(t) is the standardized value of vibration data, L(t) is the standardized value of liquid level differential pressure data, and w1, w2, w3, and w4 are weight coefficients respectively.

5. The silt interference control system for the substation water collection well according to claim 4 is characterized in that: The decision control module includes: When S(t) ≥ threshold 1, the controller starts the sludge mixer; When S(t)≤threshold 2, the controller does not start the sludge mixer.

6. The silt interference control system for the substation water collection well according to claim 1 is characterized in that: Also included is a monitoring cycle module, the monitoring cycle module is used to set a monitoring time period, and the monitoring module is used to monitor according to the time period; The monitoring cycle module includes a dynamic adjustment function, specifically: T=T0×(1+α×ΔS) Where T is the adjusted working cycle, T0 is the standard cycle, α is the adjustment coefficient, and ΔS is the rate of change of the state evaluation value.

7. Intelligent desilting and drainage device for substation water collection well, characterized by: It includes a water pump intelligent control system and a silt interference control system for a substation water collection well as described in any one of claims 1 to 6, The water pump intelligent control system includes a water pump instruction generating unit and a water pump executing unit. The water pump instruction generating unit is used to generate control instructions for the water pump execution unit; The water pump execution unit includes pump A and pump B, and the pump A and pump B are respectively slidably connected to the well wall lifting frame of the water collection well, and the vibration sensor is also used to monitor pump A and pump B; The water pump instruction generating unit and the water pump executing unit are respectively connected to a controller, and the controller is wirelessly connected to a mobile phone APP or a control center.

8. The intelligent desilting and drainage device for a water collection well in a substation according to claim 7, characterized in that: The water pump instruction generation unit includes an alternating running time balancing function, specifically: L=(T1-T2) / (T1+T2)×100% Where: T1 and T2 are the cumulative running time of the two pumps respectively, and L is the running time imbalance. If |L| is greater than the balance threshold, the system adjusts its operation so that the running time of pump A and pump B tends to be balanced.

9. The intelligent desilting and drainage device for a water collection well in a substation according to claim 7, characterized in that: It also includes an adaptive optimization control unit, which is connected to the controller. The adaptive optimization control unit includes optimizing the water pump intelligent control system using a deep reinforcement learning model; The state space S of the deep reinforcement learning model is: S = {water level, flow rate, energy consumption, vibration}; The action space A is: A = {start, stop, speed adjustment}; The reward function R is: R = w5×(energy efficiency ratio) + w6×(life expectancy) + w7×(maintenance cost) Where: w5, w6 and w7 are weight coefficients respectively.

10. The intelligent desilting and drainage device for a water collection well in a substation according to claim 7, characterized in that: It also includes a flood warning system. The flood warning system includes a collection module and a warning module. The acquisition module is used to obtain real-time rainfall and liquid level differential pressure sensor data; The early warning module is used to issue an early warning based on the acquired data; the early warning module is connected to the mobile phone APP or the control center via wireless; The early warning module includes establishing an early warning model and performing graded early warning. Among them, the early warning model is: P(flood)=σ(W·X+b) Where P(flood) is the probability of flood occurrence, σ is the standard deviation, b is the benchmark value or threshold, X is the eigenvector, and W is the weight matrix; The graded warnings are: When P(flood) ≥ the first threshold, a red warning is issued; When the second threshold ≤ P(flood) < the first threshold, an orange warning is issued; When the third threshold ≤ P(flood) < the second threshold, a yellow warning is issued.

Images